SlideShare ist ein Scribd-Unternehmen logo
1 von 98
Downloaden Sie, um offline zu lesen
SKYKOPTER
4th Year Computer and systems Eng. ~ 2009 - 2010 ( Graduation Project )AKHBAR EL YOM ACADEMY
Skykopter 2010
AKHBAR EL YOM ACADEMY
Engineering Department
Modeling of the Flight Dynamics of
a SKYKOPTER
BCs Thesis
Skykopter 2010
AKHBAR EL YOM ACADEMY
Engineering Department
4th
year Computer and Control Systems
BCs Thesis
Academic Year 2009-2010
Modeling of the flight dynamics of
a SKYKOPTER
Supervision: Dr. Mohamed M. Elkhatib
ADham Ibrahim badawi – amr Mahmoud abdel wahab - Ehab refaat boshra
Mahmoud Mohamed abd allah - Sherif farghaly hasan – sherif sami Yousef
reham adel mohamed
JULY 2010
This thesis is submitted in partial fulfillment of the requirements for the degree of
Bachelor of Science.
Skykopter 2010
ABSTRACT
Historically, helicopters with four rotors (quadrotor) have not been very common,
mainly because most of the usual payloads could be lifted using one or two rotors.
However, the quadrotor possesses some special characteristics that make it attractive.
One, of course, is the superior payload capacity. The other is the simplicity of its control
system: just by independently adjusting the speed of each rotor it is possible to control
both the attitude and the horizontal/vertical motion. This system is particularly suitable for
small UAVs, because it reduces the mechanical complexity of the rotors (saving volume
and weight) and simplifies the control algorithms required for autonomous flight.
Although much progress has been made in the field of quadrotor UAVs, it is still a great
challenge to build a quadrotor capable of fully autonomous flight. In order to be success-
ful in selecting the appropriate control algorithms it is essential to have a complete under-
standing of quadrotor flight dynamics.
This Report presents a detailed physical model to describe quadrotor flight dynamics. It is
based on a real quadrotor, the SkyKopter.
On the final part of this Report, the model is coded into Matlab/Simulink. The simulator
is then used to study open loop flight dynamics. In the future, this simulator will also be
used to test potential control algorithms.
LIST OF CONTENTS
1	Introduction....................................................................................... 17
1.1	Background............................................................................................................17
1.2	Motivation...............................................................................................................19
1.3	Contribution.........................................................................................................20
1.4	 Overview of work undertaken...................................................................24
1.5	Organization.........................................................................................................26
2	 Literature review................................................................................ 28
2.1	 Historical role of quadrotors.................................................................28
2.2	 Why Quadrotor?.................................................................................................34
2.3	 Design in literature..........................................................................................37
2.3.1	 Breguet Brothers Gyroplane............................................................37
2.3.2	 European Aeronautic Defense and Space Company............38
2.3.3	 Pennsylvania State University..........................................................38
2.3.4	 Middle East Technical University.................................................40
2.3.5	 Australian National University.....................................................41
2.3.6	 University of British Columbia Vancouver, BC, Canada.42
2.3.7	 Cornell University..................................................................................43
2.3.8	 Swiss Federal Institute of Technology.....................................44
2.3.9	 University of Technology in Compiegne, France................45
2.3.10	Stanford University...............................................................................46
2.3.11	Australian National University, Canberra, Australia......48
3	 Skykopter Design................................................................................ 51
3.1	 System Description.............................................................................................51
3.2	Components............................................................................................................52
3.2.1	 Central Plate..............................................................................................52
3.2.2	Arms...................................................................................................................53
3.2.3	Motors............................................................................................................53
3.2.4	Propellers......................................................................................................55
3.2.5	 Flight Control..........................................................................................55
3.2.6	 Brushless control OR ESC (Electrical Speed Control)...57
3.2.7	Battery.............................................................................................................57
3.2.8	 Sender (MX-16s Transmitter 35/35B MHz)..................................58
3.2.9	 Receiver (DSL-4TOP).................................................................................59
3.2.10	Bluetooth.....................................................................................................60
3.3	 Mechanical design.............................................................................................62
3.3.1	Frame................................................................................................................62
3.3.2	 Brushless motors.....................................................................................63
3.3.3	Propellers......................................................................................................65
3.4	 Electrical Design................................................................................................66
3.4.1	 Electronic Speed Control (ESC).....................................................66
3.4.2	 Pulse Width Modulation (PWM).....................................................66
3.4.3	 Controller: Atmega644 8-bit microcontroller....................69
3.4.4	 Implementation of the interfacing circuits.........................69
3.5	 Basic Control of the Skykopter................................................................71
3.5.1	Throttle.........................................................................................................71
3.5.2	Yaw.....................................................................................................................72
3.5.3	Pitch.................................................................................................................73
3.5.4	Roll...................................................................................................................73
3.6	 System Blocks.........................................................................................................75
3.6.1	 R/C Transmitter and Receiver.........................................................75
3.6.2	 Main board (flight control)...........................................................75
4	 Modeling, simulation, testing and validation............. 78
4.1	 Dynamic Simulation..........................................................................................78
4.1.1	Throttle.........................................................................................................79
4.1.2	Yaw.....................................................................................................................80
4.1.3	Roll...................................................................................................................82
4.1.4	Pitch.................................................................................................................83
4.2	 System of Differential Equations.............................................................85
4.3	 Assembly of the model......................................................................................88
4.3.1	 Mass and inertial properties.............................................................88
4.4	 Simulink model.....................................................................................................91
4.4.1	 Simplified quad-rotor Simulink model......................................91
4.4.2	 Simulink of basic control system..................................................92
4.5	 Hardware Testing...............................................................................................94
5	 Conclusion and future work................................................... 97
5.1	Conclusion.............................................................................................................97
5.2	 Future work............................................................................................................97
References..............................................................................................................................98
LIST OF SYMBOLS AND ACRONYMS
FT 		
Sum of all four rotor lift forces
FF 		 Front rotor lift force
FB 		 Back rotor lift force
FL 		 Left rotor lift force
FR 		 Right rotor lift force
Fg
		 Gravitational force
Fmax 		 Maximum force
F1 	 Vertical force produced by the first rotor
F2 		 Vertical force produced by the second rotor
F3 		 Vertical force produced by the third rotor
F4 		 Vertical force produced by the fourth rotor
u		 Resultant force
Vψ 		 Thrust variation
V θφ, 		 Yaw deviation
Vθ 		 Thrust deviation for one motor
τmax 		 Max torque
τψ 		 Yaw torque
τCW 		 Clock Wise torque
τCCW 		 Counter Clock wise
τ L 		 Left rotor torque
τ R 		 Right rotor torque
τF 		 Front rotor torque
τB 		 Back rotor torque
τ 		 Electromagnetic torque
∑τr 	 Summation Rotor torque
∑τθ 		 Summation of moments about the y-axis
K		 Motor constant
kτ 		 Scalar torque constant
k 6,5,4,3,2,1 	 Drag coefficients
kI 		 Constant of the motor to be determined
Iz 		 Mass moment of inertia about the Z-axis
Iy
		 Mass moment of inertia about the y-axis
Ix 		 Mass moment of inertia about the X-axis
∑Im 		 Summation of all point mass inertias about the z-axis
ψ 		 Yaw angular position
ψ
•
		 Yaw angular velocity
ψ
••
		 Yaw angular Acceleration
θ 		 Pitch angular position
θ
•
		 Pitch angular velocity
θ
••
		 Pitch angular acceleration
φ 		 Roll angular position•
φ 		 Roll angular velocity
φ
••
		 Roll angular Acceleration
Χ
•
		 Linear Velocity in the X-axis direction
Χ
••
		 Linear Acceleration in the X-axis direction
Υ
•
		 Linear Velocity in the Y -axis direction
Υ
••
		 Linear Acceleration in the Y-axis Direction
Ζ
•
		 Linear Velocity in the Z -axis direction
Ζ
••
		 Linear Acceleration in the Z-axis Direction
mm 		 Motor & arm mass
m		 Quadrotor mass
l 		 Length
g 		 Gravity
va 		 Voltage in the coil of the armature
vf
		 Field voltage
vaj
		 Appropriate voltage
if
		 Field current
ia 		 Current in the coil of the armature
Lf
		 Inductance of the windings of the stator
La 		 Inductance of the windings of the armature
Rf
		 Resistance of the windings of the stator
Ra 		 Resistance of the windings of the armature
ω 		 Rotational speed of the motor
Ωi 		 Speed of the motor
j 		 Subscript to identify the motor has already been included
h		 Height
P		 Pressure
Acronyms
UAVs		 Unmanned Aerial Vehicles
VTOL	 Vertical Take Off and Landing
CAD		 Computer Aided Design
UAS		 Unmanned Aircraft Systems
STOL		 Short Take Off and Landing
VTUAV	 Vertical Takeoff Unmanned Aerial Vehicle
DC		 Direct Current
GPS		 Global Positioning System
3D		 Dimension
MAV		 Micro Aerial Vehicle
QTR		 Quad Tilt Rotor
BLDC	 Brush-Less Direct Current
ESC		 Electronic Speed Control
PWM		 Pulse Width Modulation
RPM		 Rotation Per Minute
FETs		 Field Effect Transistors
PCB		 Printed Circuit Board
CMOS	 Complementary-symmetry Metal–Oxide–Semiconductor
AVR	 Is Modified Harvard architecture 8-bit
		 single chip microcontroller
RISC		 Reduced Instruction Set Computer
PPM		 Pulse Position Modulation
SDA		 Serial Data
SCL		 Serial Clock
MEMS	 Micro - Electro Mechanical Systems
IC		 Integrated Circuit
EEPROM	 Electrically Erasable Programmable Read-Only Memory
SRAM	 Static Random Access Memory
RTC		 Real Time Counter
ADC		 Analog to Digital Convertor
CCP		 Capture, Compare and PWM Module
AREF		 Analog Reference AREF
MOI		 Moment Of Inertia
COG		 Centre Of Gravity
IMU		 Inertial Measurements Unit
OEM		 Original Equipment Manufacturer
USB		 Universal Serial Bus
PANs		 Personal Area Networks
MMW	 Millimeter Wave
Skykopter 2010
Skykopter 2010
Chapter 1
Introduction
Chapter 1 INTRODUCTION
18
1 Introduction
1.1	Background
A quadrotor unmanned air vehicle has four rotors and requires no cyclic or collective
pitch. A quadrotor UAV can be highly maneuverable, has the potential to hover and to
take off, fly, and land in small areas, and can have simple control mechanisms. However,
because of its low rate damping, electronic stability augmentation is required for stable
flight.
It is controlled by changing the speed of rotation of each motor. The front and rear rotors
rotate in a counter-clockwise direction while the left and right rotors rotate in a clockwise
direction to balance the torque created by the spinning rotors.
A quadrotor has some advantages over other rotary wing UAVs. It is mechanically simple
and is controlled by only changing the speed of rotation for the four motors. Since the
yaw rate is controlled by changing motor speed, a tail rotor is not required to control yaw
rate and all thrust can be used to provide lift. A quadrotor may also be able to fly closer
to an obstacle than conventional helicopter configurations that have a large single rotor
without fear of a rotor strike. The vehicle’s dynamics are good for agility and its four ro-
tors can allow increased payload. However, the dynamics of the quadrotor can make the
vehicle difficult to control. The challenge of controlling the vehicle can be even more dif-
ficult for a small, low cost quadrotor.
Our Skykopter has onboard electronics that include a receiver for pilot input, three gyro-
scopes and a microcontroller to control the quadrotor motion, and a rechargeable battery
pack to power the skykopter’s electronics and four motors.
19
1.2	Motivation
Transmission towers serve a number of purposes, including carrying electrical transmis-
sion lines and supporting communications networks. The number of towers erected is
increasing by the thousands each year. Tower rescues can include workers involved with
building or maintaining a tower or who have been injured or have suffered a sudden ill-
ness. These rescues are hazardous in many ways. Besides the obvious danger of work-
ing at height, towers expose rescuers to hazards they may not normally encounter on the
ground.
Quadrotor can fly in high places to monitor the problem using infrared or any type of
camera to locate arcing and hot spots, it helps technicians to monitor the problem while
they are on the ground without any risky by online streaming video.
Controlling a quadrotor is very complex and virtually impossible without modern com-
puter-based control systems. The availability of sensors and high performance small size
microcontrollers has resulted in the revival of the quadrotor concept.Afew quadrotors are
currently being designed to be used as UAVs mainly for surveillance applications.
Quadrotors specifically have multiple advantages over helicopters, the most important of
which is the reduced mechanical complexity and higher safety. Advantages may be sum-
marized as follows:
•	 No gearing required between the motor and the rotor.
•	 No variable propeller pitch required for changing the angle of attack.
•	 No rotor shaft tilting required.
•	 4 smaller rotors instead of one big rotor resulting in less stored kinetic en 	
	 ergy and thus less damage in case of accidents.
•	 Minimal mechanical complexity, quadrotors require less maintenance com 	
	 pared to both helicopters and planes.
Chapter 1 INTRODUCTION
20
1.3	Contribution
As a part of the global aim, several objectives are pursued:
•	 To create a physical model of the vehicle
•	 To study the aerodynamics of the rotor
•	 To program the model using Matlab/Simulink
•	 To investigate the flight handling characteristics in open-loop
•	 To study and summarize all the published information about quadrotors in 	
	 general and quadrotor flight dynamics in particular.
Prior to starting to study in detail the flight dynamics of the quadrotor, it is necessary to
have some basic knowledge about this type of helicopter: how is it controlled, what are
its potential applications, what are its main advantages over other types of rotorcraft, etc
Even more important is to gather all the published information about quadrotor modeling
and flight dynamics. Although there are already several reports that look at these subjects
in depth, most of the information available is still scarce and incomplete. It would be con-
venient to unite all this information into a single piece of work that could later be used as
a starting point for future research.
Creating a physical model of the quadrotor
To be able to accurately predict the flight dynamics of the quadrotor, a detailed physical
model is necessary. This model should be as general as possible, making the minimum
number of assumptions and hypotheses. Nevertheless, the model will be particularized
for a specific quadrotor, the Skykopter, although it should be easily customizable to suit
21
a different type of quadrotor. The elements that constitute this model are the following:
•	 A set of equations of motion
•	 A model of the rotor
•	 A set of equations to describe the dynamics of the motors.
•	 Additionally, a simple gust model will also be included.
Most of the works published about quadrotors focus on Control issues; they use fairly
simple models to describe the Flight Dynamics of the vehicle. Although this is very con-
venient to simplify the mathematics, it can lead to the omission of important effects that
may significantly affect flight handling. An example of this is rotor modelling. Many
works consider that the rotors are rigid, that the thrust and torque coefficients are constant
and that there are no in-plane forces and moments. But nothing of this is true: the blades
are flexible and so the thrust is not parallel to the axis of the rotor; the thrust and torque
coefficients are not constant but heavily depend on the airflow through the rotor; and there
are in-plane forces and moments that may contribute to de-stabilize the vehicle.
Fortunately, this trend is already starting to reverse. Researchers are turning their eyes to
helicopter theory to seek for clues on how to produce better models of quadrotor dynam-
ics. However, application of helicopter theory to quadrotors is not straightforward. There
are many important differences between conventional helicopters and quadrotors and this
has to be addressed. For instance, conventional helicopters keep rotor speed constant.
Hence, traditional helicopter theories are optimized for the normal values of this rotor
speed. On the other hand, most quadrotors (except those which are capable of adjusting
the pitch angle of their blades) are controlled by independently modifying the speeds of
the four rotors. These speeds can be a 50% higher or lower than the value corresponding
Chapter 1 INTRODUCTION
22
to hover. Traditional helicopter theories might fail at the lower end of this range, and they
do not predict the effect on the rotor wake of constant speed changes.It will be seen that
most of the limitations of traditional helicopter theories can be circumvented. In the case
of rotor speed, by keeping far from the lower limit and by assuming quasi-steady condi-
tions during each revolution. This Research Project will try to produce a quadrotor model
with a higher level of detail than that of already existing models. It will also try to deter-
mine whether such a model is accurate enough to be used to test those controllers which
are planned to be implemented in the real vehicle.
Studying the aerodynamics of the rotor
This is a consequence of the need to produce an accurate physical model of the whole
vehicle. It is obvious that the rotors are the most important element and thus modeling
them correctly is critical. In order to do so, several theories will have to be reviewed and
numerous tests will have to be conducted at the wind tunnel. In fact, large portions of this
Report will be devoted to this subject.
Producing a Matlab/Simulink model of the quadrotor
As it has been already mentioned, the physical model of the quadrotor will be programmed
in Matlab/Simulink. The main requirements of the simulation are:
•	 Acceptable execution times
•	 Robustness
Modular architecture, to make it as flexible as possible and easy to modify
No graphics engine will be used. It is preferred to concentrate efforts on the development
the physical model.
23
Studying the open-loop flight dynamics
A quadrotor is not likely to be flown in open loop. It is very unstable; corrections have
to be made constantly and at very high frequencies. As a result, only close-loop flying is
feasible. However, it is still important to study open-loop flight dynamics, so that we can
have a better understanding of the challenges that any stabilization/control system will
have to face. It will be seen that these “challenges” include: very fast dynamics (short
time constants), coupling between attitude and linear velocity, sensitivity to pitch/roll
disturbances, etc.
Chapter 1 INTRODUCTION
24
1.4	Overview of work undertaken
In order to fulfill the objectives previously described, a clear sequence of tasks had to
be carried out. At this point, it will be enough with briefly describing the tasks that were
performed.
First of all, an extensive work of research was done in order to fulfill the first of the objec-
tives, which was to acquire a basic understanding about the characteristics of quadrotor
machines. This research gave a considerable importance to the resources available on the
Internet, since the traditional resources that contained detailed information about specific
subjects (control techniques, dynamic models …etc) but little about the basic issues, such
as the history of the quadrotor concept or its potential applications.
In the next place came the task of building the physical model of the quadrotor. This
meant writing down the equations of motion and the equations of the motors. As for the
rotor model, several options were examined. Finally, it was decided to use a model based
on those from Prouty and Young. Although the rotor model was entirely theoretical, ex-
perimental data were needed to adjust several parameters and to validate the model itself.
Another important task was to obtain the mass and inertial properties of the quadrotor. To
do so, all the elements of the Skykopter were carefully measured and weighed. Then, a
complete CAD model of the quadrotor was made. This model was later used to obtain the
position of the centre of mass and the moments and products of inertia of the complete
assembly and its sub-assemblies.
Once the physical model was ready and the mass and inertial properties had been obtained,
it was possible to build the Matlab/Simulink model of the Skykopter. This simulation did
not introduce other simplifications apart from those which were already included in the
physical model. The result was a simulation in six degrees of freedom of the quadrotor.
25
1.5	Organization
Chapter 1: Introduction. Presents an introduction to our Skykopter, it divides to 5 main
parts. First of all, a background brief is stated. Then, our motivation to carry out this
project. After that, description for our main objectives has been stated. Finally, the work
undertaken to achieve these objectives has been done.
Chapter 2: literature review; first of all historical evolution is presented from the begin-
ning of the twentieth century to the present day. Next, the reasons that led to the choice of
the thesis are explained, and a brief description of the main areas of researches with these
vehicles is made.
Chapter 3: Design, throughout which the design process of the Skykopter is described,
it starts with the definition of the objectives to be fulfilled by the aircraft. It’s followed by
the main component of our Skykopter. Then a system description has been stated. After
that, a theory of operation that describe how our Skykopter operates.
Chapter 4: Dynamics modeling and simulation. Focuses on mathematical model that
presents the dynamics equations, cad model, that helps to find the mass and inertial prop-
erties. Then, a Simulink model to produce the output (six degree of freedom). Finally the
hardware test implementation.
Chapter 5: The thesis ends with the final conclusion and suggestions for future improve-
ments.
Skykopter 2010
Chapter 2
Literature Review
Chapter 2 LITERATURE REVIEW
28
2 Literature Review	
This Chapter presents a literature review of the unmanned aerial vehicles (UAVs). It starts
with a non technical, general discussion about UAVs. Then, it presents some fundamental
definitions related to UAVs for clarification purposes. After that a comparison between a
different types of UAVs; advantages and disadvantages. Finally we stated a brief of previ-
ously designed quadrotors.
2.1 Historical role of quadrotors
This story begins in the 20th century, when Charles Richet, a French scientist and acade-
mician, built a small, un-piloted helicopter.[1] Although his attempt was not a success,
Louis Bréguet, one of Richet’s students, was inspired by his tutor’s example. Later in
1906, Louis and his brother, Jacques Bréguet began the construction of the first quadro-
tor. Louis executed many tests on airfoil shapes, proving that he had at least some basic
understanding of the requirements necessary to achieve vertical flight.
In 1907 they had finished the construction of the aircraft which was named Bréguet-
Richet Gyroplane No.1 (Figures 2.1 and 2.2), a quadrotor with propellers of 8.1meters in
diameter each, weighting 578 kg (2 pilots included) and with only one 50hp (37.3kW) in-
ternal combustion engine, which drove the rotors through a belt and pulley transmission.
Of course at that time they had no idea how they would control it, the main concern was
to ensure the aircraft would achieve vertical flight. The first attempt of flight was done in
between August and September of 1907 with witnesses saying they saw the quadrotor lift
1.5m into the air for a few moments, landing immediately afterwards. Those same wit-
nesses also mentioned the aircraft was stabilized, and perhaps even lifted by menassisting
29
on the ground. Discouraged by the lack of success of the Gyroplane No. 1, Bréguet and
his mentor continued their pursuit to build vertical flight machines and afterwards also
temporarily dedicated themselves to the development of fixed-wing aircraft, area where
they became very successful. Louis never abandoned his passion for vertical flight air-
craft and in 1932 he became one of the pioneers of helicopter development.[1]
Fig.2.1: 3D model of the Gyroplane No. 1
Fig 2.2: Bréguet-Richet Gyroplane No. 1 of 1907
Chapter 2 LITERATURE REVIEW
30
Etienne Oemichen, another engineer, also began experimenting with rotating-wing de-
signs in 1920. He designed a grand total of six different vertical lift machines. The first
model failed in lifting from the ground but Oemichen was a determined person, so he
decided to add a hydrogen-filled balloon to provide both stability and lift. His second
aircraft, the Oemichen No. 2 (Figure 2.3), had four rotors and eight propellers, supported
by a cruciform steel-tube framework layout. Five of the propellers were meant to stabi-
lize the machine laterally, another for steering and two for forward propulsion. Although
rudimentary, this machine achieved a considerable degree of stability and controllability,
having made more than a thousand test flights in the middle of that decade. It was even
possible to maintain the aircraft several minutes in the air. In the 14th of May the machine
was airborne for fourteen minutes and it flew more than a mile. But Oemichen was not
satisfied with the poor heights he was able to fly, and the next machines had only a main
rotor and two extra anti-torque rotors.
Fig 2.3: The Oemichen No.2 of 1922
The army also had an interest for vertical lift machines. In 1921, Dr. George de Bothezat
and Ivan Jerome were hired to develop one for the US Army Air Corps. The result was
a 1678kg structure with 9m arms and four 8.1m six-blade rotors (Figure 2.4). The army
contract required that the aircraft would hover at 100m high, but the best they achieved
31
was 5m. At the end of the project Bothezad demonstrated the vehicle could be quite
stable, however it was underpowered and unresponsive, among other technical problems.
Fig 2.4: Quadrotor designed by Dr. Bothezat an Ivan Jerome. This photograph was taken at take-off during its
first flight in October 1922
Later in 1956, a quadrotor helicopter prototype called “Convertawings Model A” (see
Figure 2.5) was designed both for military and civilian use. It was controlled by varying
the thrust between rotors, and its flights were a success, even in forward flight. The proj-
ect ended mainly due to the lack of demand for the aircraft.
Fig 2.5: Convertawings Model a helicopter5
Chapter 2 LITERATURE REVIEW
32
Recently there has been an increasing interest in quadrotor designs. Bell is working on a
quad tiltrotor to overcome the V-22 Ospray (see Figure 2.6), capable of carrying a large
payload, achieving high velocity and while using a short amount of space for Vertical
Take-Off and Landing (VTOL). Much of its systems come directly from the V-22 except
for the number of engines. Also, the wing structure on the new design has some improve-
ments; it has a wider wing span on the rear rotors. As a consequence, the Bell quad tilt
rotor (Figure 2.7) aims for higher performance and fuel economy.[2]
Fig 2.6: V-22 Ospray Fig 2.7: Concept of Bell’s quad tilt rotor
Another recent and famous quadrotor design is the Moller Skycar (Figure 2.8), a proto-
type for a personal VTOL “flying car”. The Skycar has four ducted fans allowing for a
safer and efficient operation at low speeds. It was a target for much criticism because the
only demonstrations of flight were hover tests with the Skycar tethered to a crane.[3] Its
inventor, Paul Moller already tried to sell the Skycar by auction without success. Nowa-
days he focuses his work on the precursor of the Skycar, the “M200G Volantor”, and a
flying saucer-style hovercraft. This later model uses eight fans controlled by a computer
and is capable of hovering up to 3 m above the ground. This limitation is imposed by the
on-board computer due to regulations of the Federal Aviation Administrations, stating
that any vehicle that flies above 3 m is regulated as an aircraft.[4]
33
Fig 2.8: Skycar during a test flight
Quadrotors are also available to the public through radio controlled toys. Some enthusi-
asts as well as researches have been developing their own quadrotor prototypes. This is
possible due to the availability of cheap electronics and lightweight resistant materials
available to the public. Be it for personal satisfaction, entertainment, military or civilian
use, quadrotors have played an important role in the evolution of aircrafts and may prove
themselves as important means of transportation in a near future.
Chapter 2 LITERATURE REVIEW
34
2.2	Why Quadrotor?
As widely known, when compared with other aerial vehicles, vertical take-off and land-
ing (VTOL) vehicle systems have specific characteristics like flying in very low altitudes
and being able to hover that make them suitable for applications that may be impossible
to complete using fixed-wing vehicles.
Different configurations of commonly used MAVs for research purposes and in industry
are shown in Table 6.1 along with related advantages and drawbacks. Table 6.1 offers a
pictorial comparison that may be used when a new design is proposed[5].
Configuration Picture Advantages Disadvantages
Fixed-wing
(AeroVironment)
- Simple mechanics
- Silent operation - No hovering
Single
(A.V de Rostyne)
- Good controllability and
maneuverability
- Complex mechanics
- Large rotor
- Long tail boom
Axial rotor
(Maryland Univ.)
- Compactness
- Simple mechanics
- Complex control
- Weak maneuverability
Coaxial rotors
(ETHZ)
- Compactness
- Simple mechanics
- Complex
aerodynamics
35
Configuration Picture Advantages Disadvantages
Tandem rotors
(Heudiasyc)
- Good controllability and
maneuverability
- No aerodynamics
Interference
- Complex mechanics
- Large size
Quadrotors
(ETHZ)
- Good controllability and
maneuverability
- No aerodynamics
Interference
- Complex mechanics
- Large size
Blimp
(EPFL)
- Low power consumption
- Auto-lift
- Large size
- Weak
Maneuverability
Hybrid
(MIT)
- Good maneuverability
- Good survivability
- Large size
- Complex design
Bird-like
(Caltech)
- Good maneuverability
- Low power Consumption
- Complex mechanics
- Complex control
Insect-like
(UC Berkeley)
- Good maneuverability
- Compactness
- Complex mechanics
- Complex control
Fish-like
(US Naval Lab)
- Multimode mobility
- Efficient Aerodynamics
- Complex control
- Weak maneuverability
Chapter 2 LITERATURE REVIEW
36
Further, Table 6.2 presents a short and not exhaustive comparison between different
VTOL vehicle concepts. It may be observed from Table 6.2 that the quadrotor and the
.coaxial helicopter are among the best configurations if used as MFRs
A B C D E F G H
Power cost 2 2 2 2 1 4 3 3
Control cost 1 1 4 2 3 3 2 1
Payload/volume 2 2 4 3 3 1 2 1
Maneuverability 4 2 2 3 3 1 3 3
Mechanics simplicity 1 3 3 1 4 4 1 1
Aerodynamics complexity 1 1 1 1 4 3 1 1
Low speed flight 4 3 4 3 4 4 2 2
High speed flight 2 4 1 2 3 1 3 3
Miniaturization 2 3 4 2 3 1 2 4
Survivability 1 3 3 1 1 3 2 3
Stationary flight 4 4 4 4 4 3 1 2
Total 24 28 32 24 33 28 22 24
,A=Single rotor, B=Axial rotor, C=Coaxial rotors, D=Tandem rotors, E=Quadrotor
.F=Blimp, G=Bird-like, H=Insect-like
37
2.2 Design in literature
2.2.1 Breguet Brothers Gyroplane
2.2.1 In 1907, the Breguet Brothers constructed the first quad-rotor named Gyroplane No.
1. The flight was a good work to show the principle of a quadrotor. In 1922, Georges de
Bothezat built a quadrotor with a rotor located at each end of a truss structure of intersect-
ing beams, placed in the shape of a cross. Experimental aircrafts X-19 and Bell X-22A
are also designed as quad-tilt rotor aircrafts. In time due to the tremendous improvements
in manufacturing techniques and innovations in metallurgical material knowledge more
precise and smaller sensors can now be produced. The Microelectromechanical Systems
(MEMs) technology now allows the production of machine components such as gears
with sizes in 10-6 meter range. Using this MEMS technology very small accelerometers,
gyros and magnetometers are also produced, which caused the production of smaller
strapdown inertial navigation systems.As a result of this improvement in technology very
small quadrotors are developed around the world such as Mesicopter (Figure 2.9) and
Hoverbot. There is also commercially available quadrotor named DraganFlyer (Figure 2.10)
		
Fig 2.9: Mesicopter				 Fig2.10: DraganFlyer
Recently, there are several different studies in the literature about quadrotors. These works
utilized different controllers, equipments and materials.
Chapter 2 LITERATURE REVIEW
38
2.2.2 European Aeronautic Defense and Space Company
Quattrocopter (Figure 2.11) is a 65 cm electrically powered VTOL with a 20 min flight
time. Its weight is 0.5 kg. Quattrocopter has flight range of 1 km. There are six inertial
sensors in its six degree of freedom MEMS inertial measurement unit (IMU). In addition
to six inertial sensors there is one GPS unit and air data sensors (gas sensors). Total mea-
surement unit weighing 65 grams, consumes less than three watts at 5 V. The motors are
detachable so that the system can be stored in a small space.
Fig 2.11: Quattrocopter
2.2.3 Pennsylvania State UniversityIn
Pennsylvania State university two different studies had been done on quadrotors. First is
a master thesis (Figure 2.12) that had been done about a quadrotor test bench. The inertial
measurement unit consists of three analog devices gyros (ADXRS150EB), one acceler-
ometer (ADXL210EB). Attitude of the quadrotor is controlled with PI control law.
Fig.2.12: Quadrotor designed in Pennsylvania State University
39
Second work done in university of Pennsylvania (Figure 2.13) utilizes DraganFlyer as
a test bed. It has external pan-tilt ground and on-board cameras in addition to the three
onboard gyroscopes. One camera placed on the ground captures the motion of five 2.5 cm
colored markers present underneath the DraganFlyer, to obtain pitch, roll and yaw angles
and the position of the quadrotor by utilizing a tracking algorithm and a conversion rou-
tine. In other words, two-camera method has been introduced for estimating the full six
degrees of freedom (DOF) pose of the helicopter. Algorithm routines ran in an off board
computer. Due to the weight limitations GPS or other accelerometers could not be add on
the system. The controller obtained the relative positions and velocities from the cameras
only.
Two methods of control are studied – one using a series of mode-based, feedback lin-
earizing controllers and the other using a back-stepping control law. The helicopter was
restricted with a tether to vertical, yaw motions and limited x and y translations. Simu-
lations performed on MATLAB-Simulink show the ability of the controller to perform
output tracking control even when there are errors on state estimates.
Fig 2.13: Quadrotor tracking with a camera
Chapter 2 LITERATURE REVIEW
40
2.2.4 Middle East Technical University
Three orthogonal piezoelectric gyro used in the system designed in Middle East Technical
University (Figure 2.14) to control the attitude of the quadrotor. The attitude controlled
with a Linear Quadratic Regulator and PD controller. Frame consists of 45 cm rectangular
aluminum profiles.
Fig 2.14: Quadrotor designed in Middle East Technical University, Turkey
2.2.5 Australian National University
The X4-Flyer developed in ANU consists of a HC-12 a single board computer, developed
at QCAT that was used as the signal conditioning system. This card uses two HC-12
processors and outputs PWM signals that control the speed drivers directly, inputs PWM
signals from an R700 JR Slimline RC receiver allowing direct plot input from a JP 3810
radio transmitter and has two separate RS232 serial channels, the first used to interface
with the inertial measurement unit (IMU) and second used as an asynchronous data linked
to the ground based computer. As an IMU the most suitable unit considered was the
EiMU embedded inertial measurement unit developed by the robotics group in QCAT,
CSIRO weighs 50-100g. Crossbow DMU-6 is also used in the prototype. It weighs 475g.
41
The rotor used is an 11’’ diameter APC-C2 2.9:1 gear system included 6’’ per revolution
pitch with a maximum trust of 700grams. An the motors are Johnson 683 500 series mo-
tors The speed controller used is MSC30 B with a weight 26g rated 30A at 12V The pilot
augmentation control system is used. A double lead compensator is used for the inner
loop. The final setup is shown in Figure 2.15.
Fig 2.15: The X4-Flyer developed in FEIT, ANU
2.2.6 University of British Columbia Vancouver, BC, Canada
Setup developed in Department of Electrical and Computer Engineering University of
British Columbia Vancouver, BC, Canada. This work focused on the nonlinear modelling
of a quad rotor UAV. An experimental system including a flying mill, a DSP system, a
programmed microprocessor and a wireless transmitter have been used to test the flight
controller. Based on the nonlinear model, an H∞ loop shaping controller is designed
for stabilization, speed, throttle and yaw control. A microprocessor, PIC16F877, is pro-
grammed to transfer the control data to a pulse width modulated signal in order to reduce
significant CPU load which otherwise would have been associated with the DS1102. This
signal is further used to control the four rotors of the Draganflyer III via a 4 channel Fu-
taba radio transmitter working in training mode.
Chapter 2 LITERATURE REVIEW
42
In order to carry out flight control experiments, an experimental rig including a custom
designed flying mill, a personal computer, dSPACE DSP board, a microprocessor pulse
modulator, a radio transmitter and the Draganflyer III was built. A picture of the flying
mill is shown in Figure 2.16. The steel base and carbon fiber boom limit the flight route
of the UAV Draganflyer III to a half sphericity of 1 meter radius.
Based on the nonlinear model, an H∞ loop shaping controller is designed for stabiliza-
tion, speed, throttle and yaw control.Aconstraint model based predictive control (MBPC)
controller is implemented for longitudinal and lateral trajectory control.
Fig 2.16: Setup developed in Department of Electrical and Computer Engineering
University of British Columbia Vancouver, BC, Canada
43
2.2.7 Cornell University
The Autonomous Flying Vehicle (AFV) project at Cornell University has been an on-
going attempt to produce a reliable autonomous hovering vehicle. In the thrust system
MaxCim motors were used. The final vehicle weighed approximately 6.22 kg. Initially an
Extended Kalman Filter was designed to handle the estimation of both the state and the
six IMU sensor bias parameters. This filter found to be cumbersome to implement, due to
extremely large and complex Jacobian terms and instead a square root implementation of
a Sigma Point Filter (SRSPF) was considered. The final picture of the quadrotor is shown
in Figure 2.17.
Fig 2.17: Quadrotor designed in Cornell University
2.2.8 Swiss Federal Institute of Technology
In the study done at Swiss Federal Institude of Technology the mechanical design, dy-
namic modelling, sensing, and control of an indoor VTOL autonomous robot OS43 is
presented. The 3 DOF are locked.From a PC and through a standard RS232 port, orders
were sent to the test bench. The RS232 to I2C module translates the serial signals to the
I2C bus motor modules. These modules integers a P.I.D regulator on a PIC16F876 micro-
controller and are capable of open or closed loop operation in position, speed or torque
Chapter 2 LITERATURE REVIEW
44
control. The MT9-B8 IMU9 estimates with a Kalman filter the 3D orientation data and
give the calibrated data of acceleration and angular velocity. It weights about 33g and
communicates at 115kbps. The captured motion from the 3D universal joint decoded to
extract absolute orientation information, by the help of the micro optical encoders in each
axis.
The cross is made with carbon rods thus vehicle, the mass of which is around 240 grams,
is light weight. The OS4 test bench has four propulsion groups, each composed of a 29g
motor including magnetic encoders, a 6g-gear box and a 6g propeller.
Before implementation on the real system, several simulations had been performed on
Matlab. The controller’s task was to stabilize the height while compensating the initial
error on the roll, pitch and yaw angles. The real system suffered from undesired but
unavoidable delays and actuator saturation. The delays were reported to be mainly due
to RS232 communications and the actuator time constant. To emulate these lacks, two
Simulink discrete-step delay blocks had been introduced in the feedback loop and on the
actuators. Saturation level depends on the chosen actuators. The experimental setup for
this study is shown in the Figure 10.
Fig 2.18: Quadrotor designed in Swiss Federal Institude of Technology
45
2.2.9 University of Technology in Compiegne, France
Quadrotor system, manufactured by the Draganfly Innovations Inc., was used as the test
base. The four control signals were transmitted by a Futaba Skysport 4 radio. The radio
and the PC (INTEL Pentium 3) were connected using data acquisition cards (ADVAN-
TECH PCL-818HG and PCL-726). The connection in the radio is directly made to the
joystick potentiometers for the collective, yaw, pitch and roll controls. In order to sim-
plify the tuning of the controller and for flight security reasons, several switches were in-
troduced in the PC-radio interface so that each control input can operate either in manual
mode or in automatic control. Therefore the control inputs that are handled manually
were selected by the pilot while the other control inputs are provided by the computer.
The Polhemus was connected via RS232 to the PC. This type of sensor was reported to
be very sensitive to electromagnetic noise and it was install as far as possible from the
electric motors and their drivers. The Draganfly III has three onboard gyros that helped
the mini-rotorcraft’s stabilization. The dynamic model of the four rotor rotorcraft was
obtained via a Lagrange approach. And the proposed controller was based on Lyapunov
analysis using a nested saturation algorithm. The picture of the setup when it was hover-
ing is shown in Figure 11.
Fig 2.19: Quadrotor designed in University of Technology in Compiegne, France
Chapter 2 LITERATURE REVIEW
46
2.2.10 Stanford University
The name of the project that is worked on in stanford university is called STARMAC.
STARMAC consists of four X4-flyer rotorcraft that can autonomously track a given way-
point trajectory. This trajectory generated by novel trajectory planning algorithms for
multi agent systems. STARMAC project aims a system fully capable of reliable autono-
mous waypoint tracking, making it a useful testbed for higher level algorithms addressing
multiple-vehicle coordination.
The base system is the off-the-shelf four-rotor helicopter called the DraganFlyer III,
which can lift approximately 113,40 grams of payload and fly for about ten minutes at
full throttle. The open-loop system is unstable and has a natural frequency of 60 Hz, mak-
ing it almost impossible for humans to fly. An existing onboard controller slows down
the system dynamics to about 5 Hz and adds damping, making it pilotable by humans. It
tracks commands for the three angular rates and thrust. An upgrade to Lithium-polymer
batteries has increased both payload and flight duration, and has greatly enchanced the
abilities of the system.
For attitude measurement, an off-the-shelf 3-D motion sensor developed by Microstrain,
the 3DM-G was used. This all in one IMU provides gyro stabilized attitude state infor-
mation at a remarkable 50 Hz. For position and velocity measurement, Trimble Lassen
LP GPS receiver was used. To improve altitude information a downward-pointing sonic
ranger (Sodar) by Acroname were used, especially for critical tasks such as take off and
landing. The Sodar has a sampling rate of 10 Hz, a range of 6 feet, and an accuracy of a
few centimeters, while the GPS computes positions at 1 Hz, and has a differential accu-
racy of about 0.5 m in the horizontal direction and 1 m in the vertical. To obtain such ac-
curacies, DGPS planned be implemented by setting up a ground station that both receives
GPS signals and broadcasts differential correction information to the flyers.
47
All of the onboard sensing is coordinated through two Microchip 40 MHz PIC micro-
controllers programmed in C. Attitude stabilization were performed on board at 50 Hz,
and any information was relayed upon request to a central base station on the ground.
Communication is via a Bluetooth Class II device that has a range of over 150 ft. The
device operates in the 2.4 GHz frequency range, and incorporates band hopping, error
correction and automatic retransmission. It is designed as a serial cable replacement and
therefore operates at a maximum bandwidth of 115.2 kbps. The communication scheme
incorporates polling and sequential transmissions, so that all flyers and the ground station
simultaneously operate on the same communication link. Therefore, the bandwidth of
115.2 kbps is divided among all flyers.
The base station on the ground performs differential GPS and waypoint tracking tasks
for all four flyers, and sends commanded attitude values to the flyers for position con-
trol. Manual flight is performed via standard joystick input to the ground station laptop.
Waypoint control of the flyers was performed using Labview on the groundstation due to
its ease of use and on the fly modification ability. Control loops have been implemented
using simple PD controllers. The system while hovering is shown in Figure 2.20.
Fig 2.20: Quadrotor designed in Stanford University
Chapter 2 LITERATURE REVIEW
48
2.2.11 Australian National University, Canberra, Australia
Current work on the X-4 Flyer aims to solve two problems: thrust and stability. X4 Flyer
has weight of 2 kg with a length of 70 cm an 11 inch diameter rotor. The electronics are
substantially the same as the Mark I, although a lighter sensor unit has replaced the origi-
nal Crossbow IMU. The control board and ‘Eimu’ IMU were built by the CSIRO ICT
Centre. The control board is a dual HC-12 microprocessor card with digital I/O. The Eimu
is a full six-axis IMU with magnetometer. It is operated in vertical gyro mode to obtain
inertial frame reference angles. There is room inside the frame for mounting the Eimu as
close to the centre of gravity as possible.
Unlike the Mark I, the Mark II incorporates simple onboard proportional-integral deriva-
tive control. The previous iteration used a slow, off-board control system connected to the
flyer by a tether. It is anticipated that the convenient aerodynamics of the X-4 had made
sophisticated control unnecessary. In conjunction with onboard power, this allowed the
flyer to be entirely self-contained. MARK II shown in figure 2.21.[5]
Fig 2.21: X-4 Flyer Mark II
49
Skykopter 2010
Chapter 3
skykopter design
Chapter 3 skykopter design
52
3 Skykopter Design
The most important target of this particular design process is to arrive at the correct set
of requirements for the aircraft, which are often summarized in a set of specifications. In
this section we will define our mission to build a quadrotor prototype suitable for indoor&
outdoor flight, as well justify the decisions and equipment chosen for achieving this pur-
pose. The specifications for our quadrotor prototype are:
•	 Overall mass not superior to 1 kg. When it comes to quadrotors, the heavier
they get, the more expensive they are. Many quadrotors used for research do not
exceed this mass. So, aiming for maximum mass of 1 kg seems to be a suitable
target;
•	 Flight autonomy between 10 and 20 minutes. There is no point in using the
quadrotor for 2 minutes and then wait a couple of hours to recharge the batter-
ies, wasting precious time;
•	 Ability to transmit live telemetry data and receive movement orders from a
ground station wirelessly, therefore avoiding the use of cables which could be-
come entangled in the aircraft and cause an accident;
•	 The quadrotor should not fly very far from ground station so there is no need
of long range telemetry hardware and the extra power requirements associated
with long range transmissions.
53
3.1 System Description
The Skykopter helicopter used in this project is the Skykopter. It differs from traditional
helicopters regarding the design as it has 4 horizontal rotors and no vertical rotors. Tra-
ditional helicopters can adjust the angle of the rotor blades as well in order to control the
helicopter, but on the Skykopter the blades have a static angle. The only variable that can
be adjusted in flight is the rotational speed of the rotors.
The reason for this is to aid the development of a model and controllers for the Skykopter,
without the risk of flying into the wall. This will of course give reduce the movements
of the Skykopter, but it will make it easier to test if the input-output relationship is as
expected.
3.2 Components
Quad-rotor crafts generally support only a light payload, as they are required to carry
the weight of the power supply, a heavy battery, onboard. Thus, weight reduction of all
components is essential in order to allow for sufficient lift force. The controls system for
such a craft is also complex, as it requires the synchronization of four individual motors.
Chapter 3 skykopter design
54
3.2.1 Central Plate
Fig 3.1: Central plate
The central plate carries all of the electronics, sensors, and battery. It sits lower than the
four motors in order to bring the center of gravity downwards for increased stability. We
manufactured it using a rapid prototyping machine considering our design for the plate,
the rapid prototyping machine was ideal because of its ability to produce relatively com-
plex details, for example the angled holes which allow for the central plate to sit lower
than the surrounding motors. The thermoplastic polymer used in rapid prototyping has
good strength to weight ratio.
3.2.2 Arms
Fig 3.2: Arms
The arms of our quad-rotor design needed to be light and strong enough to withstand the
stress and strain caused by the weight of the motors and the central plate at their opposite
ends. Carbon fiber was deemed the best choice because of its weight to strength ratio.
The thickness of the tube was chosen to be the smallest possible to lower its weight. The
55
length of each arm (10”) was chosen based on the propellers. The propellers used are 10”
long each so we had to allow enough room for them to spin without encountering turbu-
lence from one another. Since such a phenomenon would be quite complex to analyze,
we simply distanced the motors far enough apart to avoid the possibility of turbulence
interference among rotors. [7]
3.2.3 Motors
Fig 3.3: Brushless Motor
Brushless motors called out runners have become very popular with radio controlled
airplane hobbyists because of their efficiency, power, longevity and light weight in
comparison to traditional brushed motors. [8]
The 12 pole “outrunner” type design creates significant torque and can thus drive direct
props without the need for a gearbox. [9]
Outrunner are used to implement the proposed system Brushless DC motor In aircrafts
it’s all about power to weight ratio. Outrunner brushless motors produce more torque
output than conventional brushless motor. A motor can produce power up to 90 W when
it only weights 48g. The motors are cobalt, brushed, DC motors rated for 12 V, 15 A.
The DC, brushed motor configuration was desired for ease to control via PWM (Pulse-
Chapter 3 skykopter design
56
width modulation). Each one of these brushless motors uses a special electronic speed
control (ESC). The ESC’s accept a 50 Hz PWM signal with a 1 millisecond duty cycle
corresponding to stopped and a 2 millisecond duty cycle corresponding to full power. The
cobalt motors use strong rare earth magnets and provide the best power to weight ratio
of the hobby motors available for model aircraft. We were limited to these hobby motors
by our design budget. As a result, the rest of our structural design revolves around the
selection of these motors and the allowable weight of the craft based on the lift provided
by these motors.
3.2.4 Propellers
Fig 3.4: Propellers
The propellers are 10” from tip to tip. Two are of the tractor style, for clockwise rotation,
and the other two are of the pusher style, for counterclockwise rotation. For our design, a
propeller with a shallow angle of attack was necessary as it provided the vertical lift for
stable hovering. The propellers we used were steeper than the ideal design because of lim-
ited availability of propellers that are produced in both the tractor and pusher styles. [7]
The lower pitch angle lead to less torque per turn thus more spin speed.Using the selected
motors and propellers the quadrotor is capable of maximum theoretical liftoff weight of
2.8Kg. [9]
57
3.2.5 Flight Control
Fig 3.5: Flight controller
The flight control board of the model after soldering the gyros, accelerometers, pressure
sensors. The main board of the quadrotor has 3 single axis MEMS based gyros for esti-
mating the tilting degrees for each axis (Euler angles; Phi, Theta, Epsi) while the 3-axis
MEMS based accelerometer is to estimate the linear acceleration of the body and hence
we can find the velocity and the displacement. Also there is a pressure sensor for estimat-
ing the altitude of the quadrotor in flight and the main part of the board is the Atmel AT
mega microcontroller which will role all of that to achieve a stationary stable flight, one
of sensors is Linear Accelerometer which is a low-power three axes linear accelerometer
that includes a sensing element and an IC interface able to take the information from the
sensing element and to provide an analog signal to the external world.
The chip is powered by 3.3V and outputs a voltage proportional to acceleration. The ac-
celerometers can be used to determine differential position through integration. Addition-
ally, the sensor measures acceleration due to gravity and can be thus used to determine tilt
or orientation of a static object, and Gyroscope Sensor is to measure the angular velocity.
The output signal is a voltage proportional to the rate of rotation about the axis normal to
the top surface of the chip. Regular sampling of this voltage signal was required to detect
the rate of rotation of the module. By 3 sensors of this type, the angular velocities of the
X, Y and Z axis determined. These data are used to stabilize the Quad-rotor.
Chapter 3 skykopter design
58
3.2.6 Brushless control OR ESC (Electrical Speed Control)
Fig 3.6: Electrical speed control
The BL-Ctrl board is a sensor less driver for brushless DC current motors.
The ESC (electrical speed control) board is a sensor less driver for brushless DC current
motors. They are especially designed for use in quadrotor, where quick set-changes are
necessary. The control of brushless motors is triphasic in groups of PWM pulses. The
PWM frequency determines the height of the phase-voltage. The ESC’s accept a 50 Hz
PWM signal with a 1 millisecond duty cycle corresponding to stopped and a 2 millisec-
ond duty cycle corresponding to full power.
3.2.7 Battery
Fig 3.7: Battery
The battery was selected on the basis of power requirements for the selected motor/gear-
box combination. We opted for a battery of the lithium-polymer variety, despite the fact
that it was considerably more expensive than other batteries providing the same power,
because this battery provided the best power-to-weight ratio. Our battery choice was a
59
2200mah (Milliamp Hour) 12.0V 12C Li-polymer battery.
(Note: Because we did not have enough time to integrate the circuitry of the controls
system on-board, and thus performed only tethered flight, we did not ultimately pur-
chase the battery.)[7]
3.2.8 Sender (MX-16s Transmitter 35/35B MHz)
Fig 3.8: MX-16s transmitter
Microcomputer control system for 8 control functions (channels)
High-Technology micro-computer system with new high-speed single-chip micro-
computer, flash memory and 10-bit A / D Converter With advanced technology opti-
mized computer radio control system with 12 model memories. High operational safety
through modern Computer system. Easy programming through simplified programming
technique with rockers and buttons. High-contrast graphic screen provides an efficient
control of the parameters, operating modes, timers and battery voltage. Modern hard-
ware and integrated Synthesizer system for channel selection, with security against ac-
cidental start-up. Configuration and programming based on proven concepts of MC-19
MC-24.8 and control functions with a simplified method of assigning controls for auxil-
iary functions such as external switches and proportional donors provide high efficiency.
Free assignment of all switches to switched functions simply by operating the desired
Switches. 12 model memories.
Chapter 3 skykopter design
60
3.2.9 Receiver (DSL-4TOP)
Fig 3.9: DSL-4top
This receiver is a receiver for remote controlled aerial vehicles like model helicopters. Its
output signal is a sort of pulse width modulated. A single-receiver with the highest range
and selectivity Special version with multi-signal (sum-signal) output The DSL 4top is a
small 4(9)-channel micro-receiver of the company ACT.
3.2.10 Bluetooth
Fig 3.10: Bluetooth module
Is an open wireless protocol for exchanging data over short distances from fixed and
mobile devices, creating personal area networks (PANs). It was originally conceived as
a wireless alternative to RS232 (Recommended Standard 232) is a standard for serial bi-
nary single-ended data and control signals connecting data cables.
61
A Bluetooth module is mounted on the flying machine and it is attached to the main
board. A wireless connection is created between the flying machine and ground based
computer (must have Bluetooth card). Another advantage granted by Bluetooth connec-
tion it provides secured data line between the ground station and the mobile device. The
F2M03GXA is an embedded Bluetooth 2.0 EDR with excellent radio transmission char-
acteristics, highly efficient, Omni directional on-board antenna and serial port profile
(SPP). On the module is running by default, the extremely reliable and easy to use Wire-
less UART firmware that implements the Bluetooth Serial Port Profile (SPP). Any infor-
mation that is received at the serial interface is transmitted transparently via Bluetooth
to the connected remote device. The ‘pairing’ of two modules is possible. In that case a
greater range than with a simple Bluetooth dongle can be achieved (1000m).
Chapter 3 skykopter design
62
3.3 Mechanical design
3.3.1 Frame
Fig 3.11: frame
The Quadrotor consists of a base in the center and 4 arms equally spaced in a shape of a
cross. The motors are mounted on the end of the arms and connected to the rotors. The
base of the Quadrotor is the joint between the 4 arms and contains a battery and main
electrical control component.
Four DC Motors mounted at the four corners of a square, this configuration obviates the
need for a tail rotor to produce the anti-torque required to counter the main rotor torque.
One pair of diagonally opposite rotors rotates in a clockwise direction and another pair in
counter-clockwise direction, nullifying the torque on the MAV (Micro Air Vehicle) due
to the former pair.
63
3.3.2 Brushless motors.
Fig 3.12: Brushless motor
The direct current (DC) motor is one of the first machines devised to convert electrical
energy to mechanical power. Its origin can be traced to machines conceived and tested by
Michael Faraday, the experimenter who formulated the fundamental concepts of electro-
magnetism.
These concepts basically state that if a conductor, or wire, carrying current is placed in
a magnetic field, a force will act upon it. The magnitude of this force is a function of
strength of the magnetic field, the amount of current passing through the conductor and
the orientation of the magnet and conductor. The direction in which this force will act is
dependent on the direction of current and direction of the magnetic field.
As the name implies, BLDC motors do not use brushes for commutation; instead, they
are electronically commutated. BLDC motors have many advantages over brushed DC
motors and induction motors. A few of these are:
•	 Better speed versus torque properties
•	 High dynamic response
•	 High efficiency
•	 Long operating life
•	 Noiseless operation
•	 Higher speed ranges
Chapter 3 skykopter design
64
BLDC motors come in single-phase, 2-phase and 3-phase configurations. Corresponding
to its type, the stator has the same number of windings. Out of these, 3-phase motors are
the most popular and widely used. This application note focuses on 3-phase motors.
Fig 3.13: Three phase configuration
The stator of a BLDC motor consists of stacked steel laminations with windings placed in
the slots that are axially cut along the inner periphery.
Traditionally, the stator resembles that of an induction motor; however, the windings are
distributed in a different manner. Most BLDC motors have three stator windings con-
nected in star fashion. Each of these windings are constructed with numerous coils in-
terconnected to form a winding. One or more coils are placed in the slots and they are
interconnected to make a winding. Each of these windings is distributed over the stator
periphery to form an even numbers of poles.
65
3.3.3 Propellers
Fig 3.13: Pusher and puller styles of propellers
The propellers are 10 inch from tip to tip. Two are of the tractor style, for clock wise ro-
tation, and the other two are of the pusher style, for counter clock wise rotation. For our
design, a propeller with a shallow angle of attack was necessary as it provided the vertical
lift for stable hovering. The propellers we used were steeper than the ideal design because
of limited availability of propellers that are produced in both the tractor and pusher styles.
Chapter 3 skykopter design
66
3.4 Electrical Design
Skykopter is piloted with the Spectrum MX-16s which has 8 programmable channels for
signal mixing and processing. Remote pilot input is transmitted to a receiver onboard the
aircraft which is wired to a separate speed controlled unit for each of the four rotors. In
general, a speed controller functions by accepting an electrical signal from the receiver to
regulate the voltage delivered from the battery to the motor.
3.4.1 Electronic Speed Control (ESC)
The ESC board is a sensorless driver for brushless DC current motors. They are especially
designed for use in quadrotor, where quick set-changes are necessary. The control of
brushlss motors is tri-phasic in groups of PWM pulses. The PWM frequency determines
the hight of the phase-voltage (actually the arithmic mean of the voltage).
The phase voltage on the motor (as group of PWM pulses) determines the rotations per
minute (RPMs) of the motor: a motor generates a countercurrent when turning (like a
generator). the resultant RMPs will be the balance between current, coutercurrent and
power output. At any time 2 of the 6 FETs are operational to power the motor. The time of
commutation (the time at which the 2 FETs must be switched off and the next two swit-
ched on to go to the next phase), is determined by voltage measurement (actually com-
parison) of the unpowered phase of the three motor wires. To do this the different analog
comparators of the Atmega8 are used.
3.4.2 Controller: Atmega644 8-bit microcontroller
TheATmega644 is a low-power CMOS 8-bit microcontroller based on theAVR enhanced
RISC architecture. By executing powerful instructions in a single clock cycle, the ATme-
ga644 achieves throughputs approaching 1 MIPS per MHz allowing the system designed
67
to optimize power consumption versus processing speed.
One of the microcontroller’s functions and the one that it provides in the control part is
to take operator inputs (thrust, pitch, roll and yaw) form the receiver (PPM: Pulse Posi-
tion Modulation Signal) and process it then transferred to the ESCs (PWM: Pulse Width
Modulation Signal)
3.4.3 Implementation of the interfacing circuits
The receiver is connected to the Atmega644 through pin PD6, the transmitters signal is
obtained by the receiver then transferred to the microcontroller as a PPM signal. After
that the signal is processed and compared with data coming from the accelerometers and
gyroscopes (will be explained later in the chapter). Four Brushless ESCs are connected in
parallel configuration to pins PC1 and PC0 used for SDA (Serial DAta) and SCL (Serial
CLock) respectively. Data is transferred to the four ESCs. Each ESC is identified indi-
vidually using 3 nodes by connecting or disconnecting (soldering or not soldering) these
points in four different configuration. The Brushless ESCs controls the corresponding ro-
tors speed using PWM signal. The integrated system acts as a closed-loop system to give
the stability desired and maximum control on the Quad-rotor.
Fig3.15: Control Block Diagram
Receiver: Operator inputs
(Thrust, tilt and direction)
Microcontrollers
Sensors: Feed back
(Thrust, roll, yaw and pitch)
ESCs Motors
Chapter 3 skykopter design
68
3.4.4 Pulse Width Modulation (PWM)
Pulse Width Modulation, or PWM, is a technique for getting analog results with digital
means. It may be used as an efficient light dimmer or DC motor speed controller. The
circuit described here is for a general purpose device that can control DC devices which
draw up to a few amps of current. The circuit may be used in either 12 or 24 Volt systems
with only a few minor wiring changes. This device has been used to control the brightness
of an automotive tail lamp and as a motor speed control for small DC fans of the type used
in computer power supplies.
A PWM circuit works by making a square wave with a variable on-to-off ratio, the aver-
age on time may be varied from 0 to 100 percent. In this manner, a variable amount of
power is transferred to the load. The main advantage of a PWM circuit over a resistive
power controller is the efficiency, at a 50% level, the PWM will use about 50% of full
power, almost all of which is transferred to the load, a resistive controller at 50% load
power would consume about 71% of full power, 50% of the power goes to the load and
the other 21% is wasted heating the series resistor. Load efficiency is almost always a
critical factor in solar powered and other alternative energy systems.
One additional advantage of pulse width modulation is that the pulses reach the full supply
voltage and will produce more torque in a motor by being able to overcome the internal
motor resistances more easily. Finally, in a PWM circuit, common small potentiometers
may be used to control a wide variety of loads whereas large and expensive high power
variable resistors are needed for resistive controllers.
The main Disadvantages of PWM circuits are the added complexity and the possibility of
generating radio frequency interference (RFI).
69
Fig 3.14: PWM
The main use of PWM is in power electronics, for example for motor control. When gen-
erating analogue signals, the bad thing is that PWM resolution goes rapidly down with
required signal bandwidth. For example, 100 kHz PWM period could be sufficient for
generating hi quality audio signal without the need of complicated analogue output filters.
With 20 MHz base frequency you get 7.64-bit resolution, with 100 MHz you get to 10
bits. This is good for toys, but a hi-fi audio signal needs 16-bit resolution.
Chapter 3 skykopter design
70
3.5 Basic Control of the Skykopter
The Skykopter is controlled by alternating the rotational velocity of the rotors. The front
and back rotors are rotating counter clockwise and the rotors to the left and right are ro-
tating clockwise. A main board is part of the electronics on the Skykopter. It handles the
incoming control signals in the form of throttle, yaw, pitch and roll and translates them
into motor control signals. Yaw is also known as the heading. The main board makes it
possible to control the Skykopter like a traditional helicopter.
3.5.1 Throttle
When throttle is increased the revolution on the four motors increases simultaneously.
The rotors will then create a lift and the Skykopter will increase in altitude. In figure 3.16
the bars indicate that the rotational speed is above average. The Skykopter is seen from
above and the black bar is the front.
Fig 3.16: Illustration of the Skykopter with increased throttle which
results in increased altitude
71
3.5.2 Yaw
When the heading is changed the overall lift is maintained, although the speed of the ro-
tors changes. To make the Skykopter rotate clockwise the speed of the left and right ro-
tors are decreased and the speed on the front- and back rotors are increased. This way the
Skykopter maintains the same lift force. Furthermore, the Skykopter increases the drag
force from the two rotors, with the highest velocity, which makes the Skykopter rotate.
An illustration of the Skykopter turning clockwise can be seen in figure 3.17. To make it
turn counter clockwise the opposite is obviously applied.
Fig 3.17: Illustration of the change of rotor speed needed to turn the Skykopter clockwise.
Chapter 3 skykopter design
72
3.5.3 Pitch
Pitch is used in order to make the Skykopter move forward and backwards. Moving for-
ward is a result of tilting the Skykopter forward, where the velocity of the front rotor is
decreased, and the back rotor is increased. During this movement the total lift needs to be
maintained, keeping the velocity of the side rotors unchanged. Figure 3.18 illustrates the
Skykopter pitching backwards. The opposite happens when tilting the Skykopter back-
wards.
Fig 3.17: Illustration of the changing rotor speed, needed to tilt the Skykopter backwards.
73
3.5.4 Roll
Roll is used to make the Skykopter move side wards. Roll is performed like pitch, except
that the velocity of the side rotors are changed, instead of the front- and back rotors.
Fig 3.19: Illustration of change of rotor speed needed to make the Skykopter roll to the left.
Chapter 3 skykopter design
74
3.6 System Blocks
The overall system has been described, and the system is now divided into blocks, in or-
der to make an analysis. A block diagram of the system blocks is illustrated in figure 2.8.
Fig 3.20: Blocks in the system.
3.6.1 R/C Transmitter and Receiver
The first block is the R/C transmitter that sends 4 control signals (throttle, yaw, pitch and
roll). This is sent to the R/C receiver that passes the signals to the Skykopter main board.
The data is sent as pulse signals.
3.6.2 Main board (flight control)
Increasing or decreasing throttle or yaw signal will affect all the motors, while pitch and
roll only affects 2 of the motors. The input received by the Skykopter is the only informa-
tion about throttle, pitch, yaw and roll. The main board translates this information into
motor control signals, which is further described in appendix A.
75
Skykopter 2010
Chapter 4
modeling, simulation, Testing and validation
Chapter 4 MODELING AND SIMULATION
78
4 modeling and simulation
4.1 Dynamic Simulation
Dynamic attitude model was derived from Newton’s Laws. The gyroscopic Procession
of each rotor cancels out due to the counter-rotating pairs, which removes any coupling
between the pitch and roll dynamics. Due to the low rotor inertia relative to the craft’s ro-
tational inertia, the response of the electric motor was faster than the attitude dynamics, so
the motor response was neglected in this model. The total collective thrust FT is the sum
of all four rotor forces. (Subscripts are T = Total, F = Front, B = Back, L = Left, R = Right)
FFFFF RLBFT
+++= 	[10] 								(4.1)
This collective thrust is nominally equal to the gravitational force when hovering; how-
ever, it can be varied by the pilot with the throttle input up to a maximum of Fg
×2 , due
to the 2:1 thrust/weight ratio. When quad rotor is in a stationary hover, FT equals the
weight force of the entire aircraft due to gravity.
Aerodynamics, forces and moments
Fig ‎4.1: Plus configuration for flight control
79
4.1.1 Throttle
In figure 4.2 a step applied to the throttle on both the non-linear and the linear model, is
shown. Both models react like expected since the z position is increasing. It is not expect-
ed that the throttle step should change the Euler angles, which can be verified in figure
4.2. The two models keep the same angle values.
Fig 4.2: Position values when applying a step to the throttle input on the linear and non-linear model.
And Euler angle values when applying a step to the throttle input on the linear and non-linear model.
Chapter 4 MODELING AND SIMULATION
80
4.1.2 Yaw
The step applied to the yaw input is shown in figure 4.3 and makes the models change
in z values, which not affect the x and y position of the models. The positive step on the
yaw input results in a negative Euler angle on the z axis. The linear model reacts like the
non-linear model, but is more aggressive. The linear model is sufficient enough, and will
be used in designing the state-space controller.
Fig 4.3: Plot of the positions affected by a step on pitch input. And plot of the Euler angle
affected by a step on the yaw input.
4.1.2.1 Yaw dynamics
A quad rotor has two sets of counter-rotating propellers; therefore, the net yaw moment
generated from aerodynamic drag is cancelled out in neutral flight. This eliminates the
need for a tail rotor that normally wastes 12% of the power in a conventional helicopter.
Furthermore, a yaw moment is induced on a quad rotor by proportionally varying the
speeds of the counter-rotating pairs. The thrust variationVψ
is given by:
).max,42( max
max
Torquesaturationmotoravoidtoork
kV ==≤ τ
τ
ψ
[10]			 (4.2)
81
From ψτψ
×= Iz
, where Iz is the mass moment of inertia and yaw acceleration is
)( angleYawwhere
Iz
==Ψ ψ
τψ [10] (4.3)
Yaw moment is the sum of all rotor torques (CW = Clockwise, CCW = counterclockwise)
)()( ττττττττψ BFRLCCWCWr
+−+=−==∑ [10] 	 (4.4)
The magnitudes of thrust forces are set so that FF RL
= are both larger than FF BF
= . The
increased drag of the motors with higher thrust will create a net reaction moment that will
rotate the body in one yaw direction. Similarly, the body can be rotated in the opposite
yaw direction by reversing the relative magnitudes of the above pairs of thrust forces.
Note that during yaw movement of the quad rotor 0≠τψ
(net torque on the body), i.e. the
sum of reaction moments is non-zero. When the quad rotor body is not rising or dropping
in altitude, the sum of all thrusts equals the weight force due to gravity. The torque on
each rotor, caused by aerodynamic drag, is proportional to the thrust by a scalar constant
kτ
Therefore, Equation (4.4) becomes:
VkVkVkVkVk ψτψτψτψτψτψτ 4)()( =−−−+= [10] 	 (4.5)
The z-axis “Moment Of Inertia” (MOI) of the quad rotor is the sum of all point mass in-
ertias about the z-axis (assuming battery and controller inertia is negligible due to their
masses being located predominantly at the “Centre of Gravity”, or COG).
)&(4
2
massarmmotoraiswheremlmII mmmz
==∑ [10] 	 (4.6)
Therefore, substituting Equations (4.5) and (4.6) into (4.3), gives the equation of motion
for yaw acceleration:
lm
Vk
lm
Vk
I mmz
22
4
4 ψτψτψτψ === [10] (4.7)
Chapter 4 MODELING AND SIMULATION
82
4.1.3 Roll
When applying a step on the roll input the models results in positive x direction values,
shown in figure 4.4. The linear model is decreasing faster than the non-linear, which is
caused by the gyroscope effect on the non-linear model. The step on the roll input in fig-
ure 4.4 have a positive Euler angle effect around the x axis. In The figures it can be seen
that the models act as expected.
Fig 4.4: Position values when applying a step to the roll input on the linear and non-linear model.
And Euler angle values when applying a step on roll input on the linear and non-linear model.
83
4.1.4 Pitch
The step applied on the pitch input is shown in figure 4.5 and makes both models move in
the positive x direction. Both models react as expected, but as on roll step, the non-linear
is less aggressive due to the gyroscope effect. The angles shown in figure 4.5 for the roll
step, results in positive y values.
Fig 4.5: Position values when adding a pitch input step on the linear and non-linear model.
And Euler angle values when adding a pitch input step on the linear and non-linear model.
Chapter 4 MODELING AND SIMULATION
84
4.1.4.1 Pitch and roll dynamics
Due to the symmetrical nature of the quad rotor configuration, pitch and roll can be rep-
resented by the same model. Figure 4-6 illustrates the thrust variations required to induce
a moment about the y-axis for rolling. The yaw deviation limit is thus
),42( max
max
,
ForceMaximumsaturationmotoravoidtoork
k F
F
V ==≤θφ
[11] (4.8)
The equation of motion for this pitching or rolling moment is derived from the sum of
moments about the y-axis:
∑ ×= θτθ

Iy
[11] (4.9)
The thrust deviation for one motor can be calculated as
2/)( FFV FB
−=θ
[11] (4.10)
Therefore the sum of the moments is
lV∑ = θθτ 2 [11] (4.11)
The y-axis moment of inertia of quad rotor is the sum of the two point mass inertias
∑ == lmII mmy
2
2 [11] (4.12)
We now substitute Equations (4.11) and (4.12) into (4.9) to find pitch acceleration.
l
l
m
V
lm
V
mm
θθ
θ == 2
2
2
 [11] (4.13)
Due to symmetry of the quad rotor body, this also represents pitch dynamics. The dy-
namic equations discussed so far have treated the quad rotor as a flying “+” structure.
The above equations for the Quadrotor model were taken from “Advances in Unmanned
Aerial Vehicles” by (S.Bouabdallah, R. Siegwart), it is an advanced model, hence we
thought about simplifying that model.
85
4.2 System of Differential Equations
The dynamics of the quad-rotor UAV are determined from a set of equations of motion.
The complexity of the equations of motion increases with increased accuracy. The set of
equations presented model the motion of the craft based on the amount of lift delivered by
each individual motor without taking into account the aerodynamics of the craft. Motion
is obtained by varying the amount of lift each motor provides. The amount of lift each
motor provides is controlled by the amount of power delivered to each motor. Depending
on the motor, the relationship can be linear, parabolic, or a combination of various other
trigonometric functions.
For example, the craft will move in the positive x-direction by reducing the thrust from
motor A (by reducing the power) and simultaneously increasing the thrust (by increas-
ing the power) from motor C. The thrust from motor B and D must be increased so that
the craft maintains constant altitude while moving along the desired path. More complex
movements can be achieved by varying the power delivered to all four motors. The equa-
tions of motion that govern the dynamics of the craft are listed below
Linear Acceleration in the X-axis Direction:
m
u k
•
•• Χ⋅−+
=Χ 1
)cossincossin)(sin1( φθψφψ
[11]	 (4.14)
Linear Acceleration in the Y-axis Direction:
m
u k
•
•• Υ⋅−+
=Υ 2
)sincoscossin)(sin1( φψφθψ
[11]		 (4.15)
Linear Acceleration in the Z-axis Direction:
g
m
u k −
Ζ⋅−
=
•
••
Ζ 3
coscos)1( φθ
[11] 			 (4.16)
Chapter 4 MODELING AND SIMULATION
86
Pitching angular Acceleration in the X-axis Direction:
Ι






−=
•••
y
iu k θθ 5
)2( [11] 		 (4.17)
Rolling angular Acceleration in the X-axis Direction:
Ι






−=
•••
x
iu k φφ 4
)3( [11] 		 (4.18)
Yawing angular Acceleration in the X-axis Direction:
Ι






−=
•••
z
iu k ψψ 6
)4( [11] 			 (4.19)
Equations of vertical forces produced by the four rotors:
FFFFu 4321
)1( +++= [11] 			 (4.20)
FFu 13
)2( −= [11] 			 (4.21)
FFu 24
)3( −= [11] 			 (4.22)
FFFFu 4321
)4( −+−= [11] 			 (4.23)
: Vertical forces produced by four rotors
: Moments of inertia directions
: Pitch angle, roll angle, yaw angle
: Drag coefficients
FFFF 4321
,,,
III zyx
,,
ψφθ
ki ,...3,2,1=
87
4.3 Assembly of the model
The battery, camera, electronic circuits will be acting by weight forces only and the figure
below shows the assembly of the model and showing the main axes of direction that will
be considered the positive direction while studying the Simulink model of the quadrotor.
Fig 4.6: Quadrotor Assembly
Where i, j, k are the unit vectors associated to those axes.
Chapter 4 MODELING AND SIMULATION
88
4.3.1 Mass and inertial properties
4.3.1.1 Mass properties
These properties were estimated using the inspection tool of Solid edge, the mass
properties of each part is illustrated below.
Item )Mass (grams Quantity Total mass
battery 223 1 223
Square tube 26 4 104
Central plate 4 2 8
Control circuits 90 1 90
DC motor 43 4 172
ESC 6 4 24
receiver 9 1 9
propeller 7 4 28
Fasteners 30 1 30
Quadrotor 688 Total 688
4.3.1.2 Inertial properties
These properties were estimated using the inspection tool of Solid edge to be used
in the Simulink model parameters, the inertial properties of each part is illustrated
below
Axis M.M. Inertia Radii of Gyration
X kg.m^2 0.003325 mm 88.19
Y Kg.m^2 0.003263 mm 87.36
Z Kg.m^2 0.006430 mm 122.64
89
4.4 Simulink model
4.4.1 Simplified quad-rotor Simulink model
Fig 4.9: Simplified Quad-rotor Simulink Model
The simplified model of the quad-rotor consists of three subsystems:
•	 Resultant Force Diagram:
The inputs of this block diagram are the resultant forces (F1, F2, F3 and F4 of the
motors) and it calculates the sum of forces acting on the quad-rotor due to variation
of motors speeds in order to estimate the transitional and rotational motion of the
quad-rotor.
•	 Euler Angles:
This block calculates Euler angles (Phi, Theta and Epsi) as a result of variation of
forces acting on the quad-rotor.
•	 System Dynamics:
It contains the equations of the system dynamics and it calculates the position of the
quad-rotor at any given time.
Chapter 4 MODELING AND SIMULATION
90
This is a cut down model and has a clear simple mathematical modeling. Some of the pa-
rameters affecting the system were neglected such as ground effect and residual angular
speed of the motors. The simplified quad-rotor model is based on Quad-rotor dynamics
equations explained previously.
4.4.2 Simulink of basic control system
4.4.2.1 Backstepping technique
In control theory, backstepping is an analysis technique for designing stabilizing controls
for a special class of nonlinear dynamical systems. Because of the recursive structure
these systems; the designer can start the design process at the known-stable system and
“back out” new controllers that progressively stabilize each outer subsystem. The process
terminates when the final external control is reached. Hence, this process is known as
backstepping [12] and [13].
Fig 4.10: Simulink of basic control system
The block diagram in Figure 4.10 explains how the controller of the quad-rotor process
data from sensors about the quad-rotor and the surrounding environment and execute
commands applied by the operator accompanied by achieving stability of the system to
insure safe flight.
91
4.5 Hardware Testing
Before attempting to fly the quadrotor we had to perform some tests to insure that these
sensors are functional. Bluetooth communication was established between the quadrotor
and a PC. Tests were applied by manual moving or rotating the quadrotor.
The diagram shown in Figure 4.10 was obtained by manually rotating the quadrotor
around the Y axis then around the X axis followed by random movement. The figure
shows the response of the angular rotation (pitch and roll) of the physical.
Fig 4.10: Gyro Testing
Chapter 4 MODELING AND SIMULATION
92
The same procedure used in testing gyros was applied to test the accelerometer. Figure
4.11 shows the response of angular acceleration of the physical model.
Fig 4.11: Accelerometer testing
93
In order to validate the Simulink model, a comparison between the simulation model and
the physical model is made using a step input for the two systems. Figure shows the out-
put from the two systems which indicate an accepted result.
Figure 4.12: the comparison between the Simulink model and the hardware
Skykopter 2010
Chapter 6
conclusion and future work
Chapter 6 CONCLUSION
96
5 Coclusion and future work
5.1 Conclusion
The aim of our project is to have fully functional quad-rotor helicopter capable of hover
and directional hover based on operator inputs. It has high stability. Also, it has the ability
of hovering at low altitudes, the system has been heavily tailored to indoor flights .It has
the ability to roll, pitch and yaw. The estimated mass and inertial properties from the CAD
model were very close to the real model. The design and construction of the physical
structure was a success both in terms of stiffness and strength of the craft, and in terms of
weight reduction. It can carry 1000 grams additional to its weight. It’s capable of monitor-
ing the altitude and the pressure through the gyros and pressure sensors.
5.2 Future work
It can be considered as future work to optimize the control law by making feedback gain
values depend on the command inputs, Display the vehicle position and other data re-
ceived through manual radio control and video reception. Provide autonomous flying in
caves, tunnels, and urban environments demands. Put 4 distance sensors to provide com-
plete vision of its surroundings. Add heat sensors to provide the ability of mind detector.
Develop an efficient algorithm for stability analysis and stabilization. Develop a larger
quad-rotor platform than is typically used in current robotics research. Analyze flying at-
titude dynamics to allow us to tune the mechanical design for the best control sensitivity.
Provide the ability to control it through the cell phone.
97
References
1.	 Leishman, J.G., The Breguet-Richet Quad-Rotor Helicopter of 1907. Verti	
	 flite. 2001.
2.	 http://www.nationmaster.com/encyclopedia/Quadrotor. 2009.
3.	 http://en.wikipedia.org/wiki/Moller_Skycar_M400. 2009.
4.	 http://en.wikipedia.org/wiki/M200G_Volantor. 2009.
5.	 Valavanis, K.P., Advanced in Unmanned Aerial Vehicle, state of the art and 	
	 the road to autonomy. 2007.
6.	 KIVRAK, A.Ö., DESIGN OF CONTROL SYSTEMS FOR A QUADRO	
	 TOR FLIGHT VEHICLE EQUIPPED WITH INERTIAL SENSORS. 2006.
7.	 Carlo Canetta, J.C., Sevan Mehrabian, Ludguier Montejo, Hendrik Thomp	
	 son, Quad-rotor Unmanned Aerial Vehicle. 2007.
8.	 Domingues, J.M.B., Quadrotor prototype. 2009.
9.	 Rami AbouSleiman, D.K., Ermal Gjioni, and Hong Chul Yang, Unmanned 	
	 Aerial Quadrotor System. 2008.
10.	 KimonP.Valavanis, ed. Advances in Unmanned Aerial Vehicles. INTELLIG	
	 ENTSYSTEMS,CONTROL,ANDAUTOMATION: SCIENCEANDENGI	
	 NEERING, ed. S.G.Tzafestas, Springer.
11.	 kemp, Modeling of flight dynamics of a quad rotor helicopter, in School of 	
	 Engineering, Aerospace sciences, Cranfield University. Master of Science 	
	 by research
12.	 Kokotovic, P.V., The joy of feedback: nonlinear and adaptive in Control 	
	 Systems Magazine. 1992. p. 7–17.
13.	 Khalil, h.k., Nonlinear Systems 3rd ed. 2002: Prentice Hall

Weitere ähnliche Inhalte

Was ist angesagt?

Few remarkable advancement in railways around the globe
Few remarkable advancement in railways around the globeFew remarkable advancement in railways around the globe
Few remarkable advancement in railways around the globeSHASHANKKUMAR426
 
DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...
DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...
DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...ijistjournal
 
K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...
K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...
K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...Kostas Alexis
 
Lithium Ion Capacitor Simplified Simulink Model using MATLAB
Lithium Ion Capacitor Simplified Simulink Model using MATLABLithium Ion Capacitor Simplified Simulink Model using MATLAB
Lithium Ion Capacitor Simplified Simulink Model using MATLABTsuyoshi Horigome
 
Device Modeling of Li-Ion battery MATLAB/Simulink Model
Device Modeling of Li-Ion battery MATLAB/Simulink ModelDevice Modeling of Li-Ion battery MATLAB/Simulink Model
Device Modeling of Li-Ion battery MATLAB/Simulink ModelTsuyoshi Horigome
 
Quadcopter Final Report
Quadcopter Final ReportQuadcopter Final Report
Quadcopter Final ReportKevin Osman
 
How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...
How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...
How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...Tsuyoshi Horigome
 
Lead-Acid Battery Simplified Simulink Model using MATLAB
Lead-Acid Battery Simplified Simulink Model using MATLAB Lead-Acid Battery Simplified Simulink Model using MATLAB
Lead-Acid Battery Simplified Simulink Model using MATLAB Tsuyoshi Horigome
 
Lithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLABLithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLABTsuyoshi Horigome
 
Lithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLABLithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLABTsuyoshi Horigome
 
Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...
Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...
Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...Mohit Kumar
 
2009 development and implementation of a control system for a quadrotor uav
2009 development and implementation of a control system for a quadrotor uav2009 development and implementation of a control system for a quadrotor uav
2009 development and implementation of a control system for a quadrotor uavjaeaj
 
Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB
Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB
Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB Tsuyoshi Horigome
 
An235 stepper motor driving
An235   stepper motor drivingAn235   stepper motor driving
An235 stepper motor drivingRicardo Godoy
 
Modelling and control of a quad rotor robot
Modelling and control of a quad rotor robotModelling and control of a quad rotor robot
Modelling and control of a quad rotor robotPrasanna Ramamurthy
 

Was ist angesagt? (20)

04762477
0476247704762477
04762477
 
Few remarkable advancement in railways around the globe
Few remarkable advancement in railways around the globeFew remarkable advancement in railways around the globe
Few remarkable advancement in railways around the globe
 
Continuous & discontinuous conduction
Continuous & discontinuous conductionContinuous & discontinuous conduction
Continuous & discontinuous conduction
 
DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...
DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...
DUAL NEURAL NETWORK FOR ADAPTIVE SLIDING MODE CONTROL OF QUADROTOR HELICOPTER...
 
K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...
K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...
K. Alexis, A. Tzes, "Revisited Dos Samara Unmanned Aerial Vehicle: Design and...
 
Lithium Ion Capacitor Simplified Simulink Model using MATLAB
Lithium Ion Capacitor Simplified Simulink Model using MATLABLithium Ion Capacitor Simplified Simulink Model using MATLAB
Lithium Ion Capacitor Simplified Simulink Model using MATLAB
 
Device Modeling of Li-Ion battery MATLAB/Simulink Model
Device Modeling of Li-Ion battery MATLAB/Simulink ModelDevice Modeling of Li-Ion battery MATLAB/Simulink Model
Device Modeling of Li-Ion battery MATLAB/Simulink Model
 
Quadcopter Final Report
Quadcopter Final ReportQuadcopter Final Report
Quadcopter Final Report
 
Quadcopter
QuadcopterQuadcopter
Quadcopter
 
How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...
How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...
How to Design of Power Management of Hybrid Circuit(Li-Ion Battery and Li-Ion...
 
Lead-Acid Battery Simplified Simulink Model using MATLAB
Lead-Acid Battery Simplified Simulink Model using MATLAB Lead-Acid Battery Simplified Simulink Model using MATLAB
Lead-Acid Battery Simplified Simulink Model using MATLAB
 
Lithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLABLithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLAB
 
Lithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLABLithium Ion Battery Simplified Simulink Model using MATLAB
Lithium Ion Battery Simplified Simulink Model using MATLAB
 
UAV como controlar cin PID
UAV  como controlar cin PIDUAV  como controlar cin PID
UAV como controlar cin PID
 
Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...
Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...
Deployable low cost outdoor system - NORTHERN INDIA ENGINEERING COLLEGE NEW D...
 
2009 development and implementation of a control system for a quadrotor uav
2009 development and implementation of a control system for a quadrotor uav2009 development and implementation of a control system for a quadrotor uav
2009 development and implementation of a control system for a quadrotor uav
 
Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB
Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB
Nickel-Metal Hydride Battery Simplified Simulink Model using MATLAB
 
Ee6351 ELECTRICAL DRIVES AND CONTROL
Ee6351 ELECTRICAL DRIVES AND CONTROLEe6351 ELECTRICAL DRIVES AND CONTROL
Ee6351 ELECTRICAL DRIVES AND CONTROL
 
An235 stepper motor driving
An235   stepper motor drivingAn235   stepper motor driving
An235 stepper motor driving
 
Modelling and control of a quad rotor robot
Modelling and control of a quad rotor robotModelling and control of a quad rotor robot
Modelling and control of a quad rotor robot
 

Ähnlich wie Skykopter 2010

UAV Building a quadcopter project
UAV Building a quadcopter projectUAV Building a quadcopter project
UAV Building a quadcopter projecthossam gouda
 
comparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdf
comparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdfcomparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdf
comparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdfteja61850
 
Report of quadcopter
Report of quadcopterReport of quadcopter
Report of quadcopterAshish Patel
 
Control net modules
Control net modulesControl net modules
Control net modulesJairo German
 
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensorASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensorAdrià Serra Moral
 
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensorASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensorAdrià Serra Moral
 
WAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLE
WAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLEWAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLE
WAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLEijiert bestjournal
 
Project seminar quadcopter
Project seminar quadcopterProject seminar quadcopter
Project seminar quadcopterShazaan Sayeed
 
Distributed Traffic management framework
Distributed Traffic management frameworkDistributed Traffic management framework
Distributed Traffic management frameworkSaurabh Nambiar
 
system on chip for telecommand system design
system on chip for telecommand system designsystem on chip for telecommand system design
system on chip for telecommand system designRaghavendra Badager
 
A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...
A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...
A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...IRJET Journal
 
Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...
Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...
Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...Anmol Dwivedi
 
CRASH AVOIDANCE SYSTEM FOR DRONES
CRASH AVOIDANCE SYSTEM FOR DRONESCRASH AVOIDANCE SYSTEM FOR DRONES
CRASH AVOIDANCE SYSTEM FOR DRONESIRJET Journal
 

Ähnlich wie Skykopter 2010 (20)

UAV Building a quadcopter project
UAV Building a quadcopter projectUAV Building a quadcopter project
UAV Building a quadcopter project
 
comparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdf
comparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdfcomparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdf
comparison_of_scada_protocols_and_implementation_of_iec_104_and_mqtt.pdf
 
Report of quadcopter
Report of quadcopterReport of quadcopter
Report of quadcopter
 
Control net modules
Control net modulesControl net modules
Control net modules
 
Sky X Tech Report
Sky X Tech ReportSky X Tech Report
Sky X Tech Report
 
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensorASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
 
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensorASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
ASerraMoral_16MS_FlutterSuppressionFlexUAVUsingLESPSensor
 
WAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLE
WAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLEWAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLE
WAR FIELD INTELLIGENT DEFENSE FLAYING‐VEHICLE
 
ltu-cover6899158065669445093
ltu-cover6899158065669445093ltu-cover6899158065669445093
ltu-cover6899158065669445093
 
Drone
DroneDrone
Drone
 
Thesis
ThesisThesis
Thesis
 
Project seminar quadcopter
Project seminar quadcopterProject seminar quadcopter
Project seminar quadcopter
 
Fms200
Fms200 Fms200
Fms200
 
Distributed Traffic management framework
Distributed Traffic management frameworkDistributed Traffic management framework
Distributed Traffic management framework
 
system on chip for telecommand system design
system on chip for telecommand system designsystem on chip for telecommand system design
system on chip for telecommand system design
 
thesis_main
thesis_mainthesis_main
thesis_main
 
A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...
A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...
A Comparison of Closed-Loop Performance of MULTIROTOR Configurations using No...
 
Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...
Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...
Rotor Resistance Control of Wound Rotor Induction Generator (WRIG) using PSCA...
 
CRASH AVOIDANCE SYSTEM FOR DRONES
CRASH AVOIDANCE SYSTEM FOR DRONESCRASH AVOIDANCE SYSTEM FOR DRONES
CRASH AVOIDANCE SYSTEM FOR DRONES
 
RT-lab based real-time simulation of flywheel energy storage system associate...
RT-lab based real-time simulation of flywheel energy storage system associate...RT-lab based real-time simulation of flywheel energy storage system associate...
RT-lab based real-time simulation of flywheel energy storage system associate...
 

Skykopter 2010

  • 1. SKYKOPTER 4th Year Computer and systems Eng. ~ 2009 - 2010 ( Graduation Project )AKHBAR EL YOM ACADEMY
  • 3. AKHBAR EL YOM ACADEMY Engineering Department Modeling of the Flight Dynamics of a SKYKOPTER BCs Thesis
  • 5. AKHBAR EL YOM ACADEMY Engineering Department 4th year Computer and Control Systems BCs Thesis Academic Year 2009-2010 Modeling of the flight dynamics of a SKYKOPTER Supervision: Dr. Mohamed M. Elkhatib ADham Ibrahim badawi – amr Mahmoud abdel wahab - Ehab refaat boshra Mahmoud Mohamed abd allah - Sherif farghaly hasan – sherif sami Yousef reham adel mohamed JULY 2010 This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science.
  • 7. ABSTRACT Historically, helicopters with four rotors (quadrotor) have not been very common, mainly because most of the usual payloads could be lifted using one or two rotors. However, the quadrotor possesses some special characteristics that make it attractive. One, of course, is the superior payload capacity. The other is the simplicity of its control system: just by independently adjusting the speed of each rotor it is possible to control both the attitude and the horizontal/vertical motion. This system is particularly suitable for small UAVs, because it reduces the mechanical complexity of the rotors (saving volume and weight) and simplifies the control algorithms required for autonomous flight. Although much progress has been made in the field of quadrotor UAVs, it is still a great challenge to build a quadrotor capable of fully autonomous flight. In order to be success- ful in selecting the appropriate control algorithms it is essential to have a complete under- standing of quadrotor flight dynamics. This Report presents a detailed physical model to describe quadrotor flight dynamics. It is based on a real quadrotor, the SkyKopter. On the final part of this Report, the model is coded into Matlab/Simulink. The simulator is then used to study open loop flight dynamics. In the future, this simulator will also be used to test potential control algorithms.
  • 8. LIST OF CONTENTS 1 Introduction....................................................................................... 17 1.1 Background............................................................................................................17 1.2 Motivation...............................................................................................................19 1.3 Contribution.........................................................................................................20 1.4 Overview of work undertaken...................................................................24 1.5 Organization.........................................................................................................26 2 Literature review................................................................................ 28 2.1 Historical role of quadrotors.................................................................28 2.2 Why Quadrotor?.................................................................................................34 2.3 Design in literature..........................................................................................37 2.3.1 Breguet Brothers Gyroplane............................................................37 2.3.2 European Aeronautic Defense and Space Company............38 2.3.3 Pennsylvania State University..........................................................38 2.3.4 Middle East Technical University.................................................40 2.3.5 Australian National University.....................................................41 2.3.6 University of British Columbia Vancouver, BC, Canada.42 2.3.7 Cornell University..................................................................................43 2.3.8 Swiss Federal Institute of Technology.....................................44 2.3.9 University of Technology in Compiegne, France................45 2.3.10 Stanford University...............................................................................46 2.3.11 Australian National University, Canberra, Australia......48 3 Skykopter Design................................................................................ 51 3.1 System Description.............................................................................................51 3.2 Components............................................................................................................52 3.2.1 Central Plate..............................................................................................52 3.2.2 Arms...................................................................................................................53 3.2.3 Motors............................................................................................................53
  • 9. 3.2.4 Propellers......................................................................................................55 3.2.5 Flight Control..........................................................................................55 3.2.6 Brushless control OR ESC (Electrical Speed Control)...57 3.2.7 Battery.............................................................................................................57 3.2.8 Sender (MX-16s Transmitter 35/35B MHz)..................................58 3.2.9 Receiver (DSL-4TOP).................................................................................59 3.2.10 Bluetooth.....................................................................................................60 3.3 Mechanical design.............................................................................................62 3.3.1 Frame................................................................................................................62 3.3.2 Brushless motors.....................................................................................63 3.3.3 Propellers......................................................................................................65 3.4 Electrical Design................................................................................................66 3.4.1 Electronic Speed Control (ESC).....................................................66 3.4.2 Pulse Width Modulation (PWM).....................................................66 3.4.3 Controller: Atmega644 8-bit microcontroller....................69 3.4.4 Implementation of the interfacing circuits.........................69 3.5 Basic Control of the Skykopter................................................................71 3.5.1 Throttle.........................................................................................................71 3.5.2 Yaw.....................................................................................................................72 3.5.3 Pitch.................................................................................................................73 3.5.4 Roll...................................................................................................................73 3.6 System Blocks.........................................................................................................75 3.6.1 R/C Transmitter and Receiver.........................................................75 3.6.2 Main board (flight control)...........................................................75 4 Modeling, simulation, testing and validation............. 78 4.1 Dynamic Simulation..........................................................................................78 4.1.1 Throttle.........................................................................................................79
  • 10. 4.1.2 Yaw.....................................................................................................................80 4.1.3 Roll...................................................................................................................82 4.1.4 Pitch.................................................................................................................83 4.2 System of Differential Equations.............................................................85 4.3 Assembly of the model......................................................................................88 4.3.1 Mass and inertial properties.............................................................88 4.4 Simulink model.....................................................................................................91 4.4.1 Simplified quad-rotor Simulink model......................................91 4.4.2 Simulink of basic control system..................................................92 4.5 Hardware Testing...............................................................................................94 5 Conclusion and future work................................................... 97 5.1 Conclusion.............................................................................................................97 5.2 Future work............................................................................................................97 References..............................................................................................................................98
  • 11. LIST OF SYMBOLS AND ACRONYMS FT Sum of all four rotor lift forces FF Front rotor lift force FB Back rotor lift force FL Left rotor lift force FR Right rotor lift force Fg Gravitational force Fmax Maximum force F1 Vertical force produced by the first rotor F2 Vertical force produced by the second rotor F3 Vertical force produced by the third rotor F4 Vertical force produced by the fourth rotor u Resultant force Vψ Thrust variation V θφ, Yaw deviation Vθ Thrust deviation for one motor τmax Max torque τψ Yaw torque τCW Clock Wise torque τCCW Counter Clock wise τ L Left rotor torque τ R Right rotor torque τF Front rotor torque τB Back rotor torque τ Electromagnetic torque ∑τr Summation Rotor torque ∑τθ Summation of moments about the y-axis
  • 12. K Motor constant kτ Scalar torque constant k 6,5,4,3,2,1 Drag coefficients kI Constant of the motor to be determined Iz Mass moment of inertia about the Z-axis Iy Mass moment of inertia about the y-axis Ix Mass moment of inertia about the X-axis ∑Im Summation of all point mass inertias about the z-axis ψ Yaw angular position ψ • Yaw angular velocity ψ •• Yaw angular Acceleration θ Pitch angular position θ • Pitch angular velocity θ •• Pitch angular acceleration φ Roll angular position• φ Roll angular velocity φ •• Roll angular Acceleration Χ • Linear Velocity in the X-axis direction Χ •• Linear Acceleration in the X-axis direction Υ • Linear Velocity in the Y -axis direction Υ •• Linear Acceleration in the Y-axis Direction Ζ • Linear Velocity in the Z -axis direction Ζ •• Linear Acceleration in the Z-axis Direction mm Motor & arm mass m Quadrotor mass l Length g Gravity va Voltage in the coil of the armature
  • 13. vf Field voltage vaj Appropriate voltage if Field current ia Current in the coil of the armature Lf Inductance of the windings of the stator La Inductance of the windings of the armature Rf Resistance of the windings of the stator Ra Resistance of the windings of the armature ω Rotational speed of the motor Ωi Speed of the motor j Subscript to identify the motor has already been included h Height P Pressure Acronyms UAVs Unmanned Aerial Vehicles VTOL Vertical Take Off and Landing CAD Computer Aided Design UAS Unmanned Aircraft Systems STOL Short Take Off and Landing VTUAV Vertical Takeoff Unmanned Aerial Vehicle DC Direct Current GPS Global Positioning System 3D Dimension MAV Micro Aerial Vehicle QTR Quad Tilt Rotor BLDC Brush-Less Direct Current
  • 14. ESC Electronic Speed Control PWM Pulse Width Modulation RPM Rotation Per Minute FETs Field Effect Transistors PCB Printed Circuit Board CMOS Complementary-symmetry Metal–Oxide–Semiconductor AVR Is Modified Harvard architecture 8-bit single chip microcontroller RISC Reduced Instruction Set Computer PPM Pulse Position Modulation SDA Serial Data SCL Serial Clock MEMS Micro - Electro Mechanical Systems IC Integrated Circuit EEPROM Electrically Erasable Programmable Read-Only Memory SRAM Static Random Access Memory RTC Real Time Counter ADC Analog to Digital Convertor CCP Capture, Compare and PWM Module AREF Analog Reference AREF MOI Moment Of Inertia COG Centre Of Gravity IMU Inertial Measurements Unit OEM Original Equipment Manufacturer USB Universal Serial Bus PANs Personal Area Networks MMW Millimeter Wave
  • 18. Chapter 1 INTRODUCTION 18 1 Introduction 1.1 Background A quadrotor unmanned air vehicle has four rotors and requires no cyclic or collective pitch. A quadrotor UAV can be highly maneuverable, has the potential to hover and to take off, fly, and land in small areas, and can have simple control mechanisms. However, because of its low rate damping, electronic stability augmentation is required for stable flight. It is controlled by changing the speed of rotation of each motor. The front and rear rotors rotate in a counter-clockwise direction while the left and right rotors rotate in a clockwise direction to balance the torque created by the spinning rotors. A quadrotor has some advantages over other rotary wing UAVs. It is mechanically simple and is controlled by only changing the speed of rotation for the four motors. Since the yaw rate is controlled by changing motor speed, a tail rotor is not required to control yaw rate and all thrust can be used to provide lift. A quadrotor may also be able to fly closer to an obstacle than conventional helicopter configurations that have a large single rotor without fear of a rotor strike. The vehicle’s dynamics are good for agility and its four ro- tors can allow increased payload. However, the dynamics of the quadrotor can make the vehicle difficult to control. The challenge of controlling the vehicle can be even more dif- ficult for a small, low cost quadrotor. Our Skykopter has onboard electronics that include a receiver for pilot input, three gyro- scopes and a microcontroller to control the quadrotor motion, and a rechargeable battery pack to power the skykopter’s electronics and four motors.
  • 19. 19 1.2 Motivation Transmission towers serve a number of purposes, including carrying electrical transmis- sion lines and supporting communications networks. The number of towers erected is increasing by the thousands each year. Tower rescues can include workers involved with building or maintaining a tower or who have been injured or have suffered a sudden ill- ness. These rescues are hazardous in many ways. Besides the obvious danger of work- ing at height, towers expose rescuers to hazards they may not normally encounter on the ground. Quadrotor can fly in high places to monitor the problem using infrared or any type of camera to locate arcing and hot spots, it helps technicians to monitor the problem while they are on the ground without any risky by online streaming video. Controlling a quadrotor is very complex and virtually impossible without modern com- puter-based control systems. The availability of sensors and high performance small size microcontrollers has resulted in the revival of the quadrotor concept.Afew quadrotors are currently being designed to be used as UAVs mainly for surveillance applications. Quadrotors specifically have multiple advantages over helicopters, the most important of which is the reduced mechanical complexity and higher safety. Advantages may be sum- marized as follows: • No gearing required between the motor and the rotor. • No variable propeller pitch required for changing the angle of attack. • No rotor shaft tilting required. • 4 smaller rotors instead of one big rotor resulting in less stored kinetic en ergy and thus less damage in case of accidents. • Minimal mechanical complexity, quadrotors require less maintenance com pared to both helicopters and planes.
  • 20. Chapter 1 INTRODUCTION 20 1.3 Contribution As a part of the global aim, several objectives are pursued: • To create a physical model of the vehicle • To study the aerodynamics of the rotor • To program the model using Matlab/Simulink • To investigate the flight handling characteristics in open-loop • To study and summarize all the published information about quadrotors in general and quadrotor flight dynamics in particular. Prior to starting to study in detail the flight dynamics of the quadrotor, it is necessary to have some basic knowledge about this type of helicopter: how is it controlled, what are its potential applications, what are its main advantages over other types of rotorcraft, etc Even more important is to gather all the published information about quadrotor modeling and flight dynamics. Although there are already several reports that look at these subjects in depth, most of the information available is still scarce and incomplete. It would be con- venient to unite all this information into a single piece of work that could later be used as a starting point for future research. Creating a physical model of the quadrotor To be able to accurately predict the flight dynamics of the quadrotor, a detailed physical model is necessary. This model should be as general as possible, making the minimum number of assumptions and hypotheses. Nevertheless, the model will be particularized for a specific quadrotor, the Skykopter, although it should be easily customizable to suit
  • 21. 21 a different type of quadrotor. The elements that constitute this model are the following: • A set of equations of motion • A model of the rotor • A set of equations to describe the dynamics of the motors. • Additionally, a simple gust model will also be included. Most of the works published about quadrotors focus on Control issues; they use fairly simple models to describe the Flight Dynamics of the vehicle. Although this is very con- venient to simplify the mathematics, it can lead to the omission of important effects that may significantly affect flight handling. An example of this is rotor modelling. Many works consider that the rotors are rigid, that the thrust and torque coefficients are constant and that there are no in-plane forces and moments. But nothing of this is true: the blades are flexible and so the thrust is not parallel to the axis of the rotor; the thrust and torque coefficients are not constant but heavily depend on the airflow through the rotor; and there are in-plane forces and moments that may contribute to de-stabilize the vehicle. Fortunately, this trend is already starting to reverse. Researchers are turning their eyes to helicopter theory to seek for clues on how to produce better models of quadrotor dynam- ics. However, application of helicopter theory to quadrotors is not straightforward. There are many important differences between conventional helicopters and quadrotors and this has to be addressed. For instance, conventional helicopters keep rotor speed constant. Hence, traditional helicopter theories are optimized for the normal values of this rotor speed. On the other hand, most quadrotors (except those which are capable of adjusting the pitch angle of their blades) are controlled by independently modifying the speeds of the four rotors. These speeds can be a 50% higher or lower than the value corresponding
  • 22. Chapter 1 INTRODUCTION 22 to hover. Traditional helicopter theories might fail at the lower end of this range, and they do not predict the effect on the rotor wake of constant speed changes.It will be seen that most of the limitations of traditional helicopter theories can be circumvented. In the case of rotor speed, by keeping far from the lower limit and by assuming quasi-steady condi- tions during each revolution. This Research Project will try to produce a quadrotor model with a higher level of detail than that of already existing models. It will also try to deter- mine whether such a model is accurate enough to be used to test those controllers which are planned to be implemented in the real vehicle. Studying the aerodynamics of the rotor This is a consequence of the need to produce an accurate physical model of the whole vehicle. It is obvious that the rotors are the most important element and thus modeling them correctly is critical. In order to do so, several theories will have to be reviewed and numerous tests will have to be conducted at the wind tunnel. In fact, large portions of this Report will be devoted to this subject. Producing a Matlab/Simulink model of the quadrotor As it has been already mentioned, the physical model of the quadrotor will be programmed in Matlab/Simulink. The main requirements of the simulation are: • Acceptable execution times • Robustness Modular architecture, to make it as flexible as possible and easy to modify No graphics engine will be used. It is preferred to concentrate efforts on the development the physical model.
  • 23. 23 Studying the open-loop flight dynamics A quadrotor is not likely to be flown in open loop. It is very unstable; corrections have to be made constantly and at very high frequencies. As a result, only close-loop flying is feasible. However, it is still important to study open-loop flight dynamics, so that we can have a better understanding of the challenges that any stabilization/control system will have to face. It will be seen that these “challenges” include: very fast dynamics (short time constants), coupling between attitude and linear velocity, sensitivity to pitch/roll disturbances, etc.
  • 24. Chapter 1 INTRODUCTION 24 1.4 Overview of work undertaken In order to fulfill the objectives previously described, a clear sequence of tasks had to be carried out. At this point, it will be enough with briefly describing the tasks that were performed. First of all, an extensive work of research was done in order to fulfill the first of the objec- tives, which was to acquire a basic understanding about the characteristics of quadrotor machines. This research gave a considerable importance to the resources available on the Internet, since the traditional resources that contained detailed information about specific subjects (control techniques, dynamic models …etc) but little about the basic issues, such as the history of the quadrotor concept or its potential applications. In the next place came the task of building the physical model of the quadrotor. This meant writing down the equations of motion and the equations of the motors. As for the rotor model, several options were examined. Finally, it was decided to use a model based on those from Prouty and Young. Although the rotor model was entirely theoretical, ex- perimental data were needed to adjust several parameters and to validate the model itself. Another important task was to obtain the mass and inertial properties of the quadrotor. To do so, all the elements of the Skykopter were carefully measured and weighed. Then, a complete CAD model of the quadrotor was made. This model was later used to obtain the position of the centre of mass and the moments and products of inertia of the complete assembly and its sub-assemblies. Once the physical model was ready and the mass and inertial properties had been obtained, it was possible to build the Matlab/Simulink model of the Skykopter. This simulation did not introduce other simplifications apart from those which were already included in the physical model. The result was a simulation in six degrees of freedom of the quadrotor.
  • 25. 25 1.5 Organization Chapter 1: Introduction. Presents an introduction to our Skykopter, it divides to 5 main parts. First of all, a background brief is stated. Then, our motivation to carry out this project. After that, description for our main objectives has been stated. Finally, the work undertaken to achieve these objectives has been done. Chapter 2: literature review; first of all historical evolution is presented from the begin- ning of the twentieth century to the present day. Next, the reasons that led to the choice of the thesis are explained, and a brief description of the main areas of researches with these vehicles is made. Chapter 3: Design, throughout which the design process of the Skykopter is described, it starts with the definition of the objectives to be fulfilled by the aircraft. It’s followed by the main component of our Skykopter. Then a system description has been stated. After that, a theory of operation that describe how our Skykopter operates. Chapter 4: Dynamics modeling and simulation. Focuses on mathematical model that presents the dynamics equations, cad model, that helps to find the mass and inertial prop- erties. Then, a Simulink model to produce the output (six degree of freedom). Finally the hardware test implementation. Chapter 5: The thesis ends with the final conclusion and suggestions for future improve- ments.
  • 28. Chapter 2 LITERATURE REVIEW 28 2 Literature Review This Chapter presents a literature review of the unmanned aerial vehicles (UAVs). It starts with a non technical, general discussion about UAVs. Then, it presents some fundamental definitions related to UAVs for clarification purposes. After that a comparison between a different types of UAVs; advantages and disadvantages. Finally we stated a brief of previ- ously designed quadrotors. 2.1 Historical role of quadrotors This story begins in the 20th century, when Charles Richet, a French scientist and acade- mician, built a small, un-piloted helicopter.[1] Although his attempt was not a success, Louis Bréguet, one of Richet’s students, was inspired by his tutor’s example. Later in 1906, Louis and his brother, Jacques Bréguet began the construction of the first quadro- tor. Louis executed many tests on airfoil shapes, proving that he had at least some basic understanding of the requirements necessary to achieve vertical flight. In 1907 they had finished the construction of the aircraft which was named Bréguet- Richet Gyroplane No.1 (Figures 2.1 and 2.2), a quadrotor with propellers of 8.1meters in diameter each, weighting 578 kg (2 pilots included) and with only one 50hp (37.3kW) in- ternal combustion engine, which drove the rotors through a belt and pulley transmission. Of course at that time they had no idea how they would control it, the main concern was to ensure the aircraft would achieve vertical flight. The first attempt of flight was done in between August and September of 1907 with witnesses saying they saw the quadrotor lift 1.5m into the air for a few moments, landing immediately afterwards. Those same wit- nesses also mentioned the aircraft was stabilized, and perhaps even lifted by menassisting
  • 29. 29 on the ground. Discouraged by the lack of success of the Gyroplane No. 1, Bréguet and his mentor continued their pursuit to build vertical flight machines and afterwards also temporarily dedicated themselves to the development of fixed-wing aircraft, area where they became very successful. Louis never abandoned his passion for vertical flight air- craft and in 1932 he became one of the pioneers of helicopter development.[1] Fig.2.1: 3D model of the Gyroplane No. 1 Fig 2.2: Bréguet-Richet Gyroplane No. 1 of 1907
  • 30. Chapter 2 LITERATURE REVIEW 30 Etienne Oemichen, another engineer, also began experimenting with rotating-wing de- signs in 1920. He designed a grand total of six different vertical lift machines. The first model failed in lifting from the ground but Oemichen was a determined person, so he decided to add a hydrogen-filled balloon to provide both stability and lift. His second aircraft, the Oemichen No. 2 (Figure 2.3), had four rotors and eight propellers, supported by a cruciform steel-tube framework layout. Five of the propellers were meant to stabi- lize the machine laterally, another for steering and two for forward propulsion. Although rudimentary, this machine achieved a considerable degree of stability and controllability, having made more than a thousand test flights in the middle of that decade. It was even possible to maintain the aircraft several minutes in the air. In the 14th of May the machine was airborne for fourteen minutes and it flew more than a mile. But Oemichen was not satisfied with the poor heights he was able to fly, and the next machines had only a main rotor and two extra anti-torque rotors. Fig 2.3: The Oemichen No.2 of 1922 The army also had an interest for vertical lift machines. In 1921, Dr. George de Bothezat and Ivan Jerome were hired to develop one for the US Army Air Corps. The result was a 1678kg structure with 9m arms and four 8.1m six-blade rotors (Figure 2.4). The army contract required that the aircraft would hover at 100m high, but the best they achieved
  • 31. 31 was 5m. At the end of the project Bothezad demonstrated the vehicle could be quite stable, however it was underpowered and unresponsive, among other technical problems. Fig 2.4: Quadrotor designed by Dr. Bothezat an Ivan Jerome. This photograph was taken at take-off during its first flight in October 1922 Later in 1956, a quadrotor helicopter prototype called “Convertawings Model A” (see Figure 2.5) was designed both for military and civilian use. It was controlled by varying the thrust between rotors, and its flights were a success, even in forward flight. The proj- ect ended mainly due to the lack of demand for the aircraft. Fig 2.5: Convertawings Model a helicopter5
  • 32. Chapter 2 LITERATURE REVIEW 32 Recently there has been an increasing interest in quadrotor designs. Bell is working on a quad tiltrotor to overcome the V-22 Ospray (see Figure 2.6), capable of carrying a large payload, achieving high velocity and while using a short amount of space for Vertical Take-Off and Landing (VTOL). Much of its systems come directly from the V-22 except for the number of engines. Also, the wing structure on the new design has some improve- ments; it has a wider wing span on the rear rotors. As a consequence, the Bell quad tilt rotor (Figure 2.7) aims for higher performance and fuel economy.[2] Fig 2.6: V-22 Ospray Fig 2.7: Concept of Bell’s quad tilt rotor Another recent and famous quadrotor design is the Moller Skycar (Figure 2.8), a proto- type for a personal VTOL “flying car”. The Skycar has four ducted fans allowing for a safer and efficient operation at low speeds. It was a target for much criticism because the only demonstrations of flight were hover tests with the Skycar tethered to a crane.[3] Its inventor, Paul Moller already tried to sell the Skycar by auction without success. Nowa- days he focuses his work on the precursor of the Skycar, the “M200G Volantor”, and a flying saucer-style hovercraft. This later model uses eight fans controlled by a computer and is capable of hovering up to 3 m above the ground. This limitation is imposed by the on-board computer due to regulations of the Federal Aviation Administrations, stating that any vehicle that flies above 3 m is regulated as an aircraft.[4]
  • 33. 33 Fig 2.8: Skycar during a test flight Quadrotors are also available to the public through radio controlled toys. Some enthusi- asts as well as researches have been developing their own quadrotor prototypes. This is possible due to the availability of cheap electronics and lightweight resistant materials available to the public. Be it for personal satisfaction, entertainment, military or civilian use, quadrotors have played an important role in the evolution of aircrafts and may prove themselves as important means of transportation in a near future.
  • 34. Chapter 2 LITERATURE REVIEW 34 2.2 Why Quadrotor? As widely known, when compared with other aerial vehicles, vertical take-off and land- ing (VTOL) vehicle systems have specific characteristics like flying in very low altitudes and being able to hover that make them suitable for applications that may be impossible to complete using fixed-wing vehicles. Different configurations of commonly used MAVs for research purposes and in industry are shown in Table 6.1 along with related advantages and drawbacks. Table 6.1 offers a pictorial comparison that may be used when a new design is proposed[5]. Configuration Picture Advantages Disadvantages Fixed-wing (AeroVironment) - Simple mechanics - Silent operation - No hovering Single (A.V de Rostyne) - Good controllability and maneuverability - Complex mechanics - Large rotor - Long tail boom Axial rotor (Maryland Univ.) - Compactness - Simple mechanics - Complex control - Weak maneuverability Coaxial rotors (ETHZ) - Compactness - Simple mechanics - Complex aerodynamics
  • 35. 35 Configuration Picture Advantages Disadvantages Tandem rotors (Heudiasyc) - Good controllability and maneuverability - No aerodynamics Interference - Complex mechanics - Large size Quadrotors (ETHZ) - Good controllability and maneuverability - No aerodynamics Interference - Complex mechanics - Large size Blimp (EPFL) - Low power consumption - Auto-lift - Large size - Weak Maneuverability Hybrid (MIT) - Good maneuverability - Good survivability - Large size - Complex design Bird-like (Caltech) - Good maneuverability - Low power Consumption - Complex mechanics - Complex control Insect-like (UC Berkeley) - Good maneuverability - Compactness - Complex mechanics - Complex control Fish-like (US Naval Lab) - Multimode mobility - Efficient Aerodynamics - Complex control - Weak maneuverability
  • 36. Chapter 2 LITERATURE REVIEW 36 Further, Table 6.2 presents a short and not exhaustive comparison between different VTOL vehicle concepts. It may be observed from Table 6.2 that the quadrotor and the .coaxial helicopter are among the best configurations if used as MFRs A B C D E F G H Power cost 2 2 2 2 1 4 3 3 Control cost 1 1 4 2 3 3 2 1 Payload/volume 2 2 4 3 3 1 2 1 Maneuverability 4 2 2 3 3 1 3 3 Mechanics simplicity 1 3 3 1 4 4 1 1 Aerodynamics complexity 1 1 1 1 4 3 1 1 Low speed flight 4 3 4 3 4 4 2 2 High speed flight 2 4 1 2 3 1 3 3 Miniaturization 2 3 4 2 3 1 2 4 Survivability 1 3 3 1 1 3 2 3 Stationary flight 4 4 4 4 4 3 1 2 Total 24 28 32 24 33 28 22 24 ,A=Single rotor, B=Axial rotor, C=Coaxial rotors, D=Tandem rotors, E=Quadrotor .F=Blimp, G=Bird-like, H=Insect-like
  • 37. 37 2.2 Design in literature 2.2.1 Breguet Brothers Gyroplane 2.2.1 In 1907, the Breguet Brothers constructed the first quad-rotor named Gyroplane No. 1. The flight was a good work to show the principle of a quadrotor. In 1922, Georges de Bothezat built a quadrotor with a rotor located at each end of a truss structure of intersect- ing beams, placed in the shape of a cross. Experimental aircrafts X-19 and Bell X-22A are also designed as quad-tilt rotor aircrafts. In time due to the tremendous improvements in manufacturing techniques and innovations in metallurgical material knowledge more precise and smaller sensors can now be produced. The Microelectromechanical Systems (MEMs) technology now allows the production of machine components such as gears with sizes in 10-6 meter range. Using this MEMS technology very small accelerometers, gyros and magnetometers are also produced, which caused the production of smaller strapdown inertial navigation systems.As a result of this improvement in technology very small quadrotors are developed around the world such as Mesicopter (Figure 2.9) and Hoverbot. There is also commercially available quadrotor named DraganFlyer (Figure 2.10) Fig 2.9: Mesicopter Fig2.10: DraganFlyer Recently, there are several different studies in the literature about quadrotors. These works utilized different controllers, equipments and materials.
  • 38. Chapter 2 LITERATURE REVIEW 38 2.2.2 European Aeronautic Defense and Space Company Quattrocopter (Figure 2.11) is a 65 cm electrically powered VTOL with a 20 min flight time. Its weight is 0.5 kg. Quattrocopter has flight range of 1 km. There are six inertial sensors in its six degree of freedom MEMS inertial measurement unit (IMU). In addition to six inertial sensors there is one GPS unit and air data sensors (gas sensors). Total mea- surement unit weighing 65 grams, consumes less than three watts at 5 V. The motors are detachable so that the system can be stored in a small space. Fig 2.11: Quattrocopter 2.2.3 Pennsylvania State UniversityIn Pennsylvania State university two different studies had been done on quadrotors. First is a master thesis (Figure 2.12) that had been done about a quadrotor test bench. The inertial measurement unit consists of three analog devices gyros (ADXRS150EB), one acceler- ometer (ADXL210EB). Attitude of the quadrotor is controlled with PI control law. Fig.2.12: Quadrotor designed in Pennsylvania State University
  • 39. 39 Second work done in university of Pennsylvania (Figure 2.13) utilizes DraganFlyer as a test bed. It has external pan-tilt ground and on-board cameras in addition to the three onboard gyroscopes. One camera placed on the ground captures the motion of five 2.5 cm colored markers present underneath the DraganFlyer, to obtain pitch, roll and yaw angles and the position of the quadrotor by utilizing a tracking algorithm and a conversion rou- tine. In other words, two-camera method has been introduced for estimating the full six degrees of freedom (DOF) pose of the helicopter. Algorithm routines ran in an off board computer. Due to the weight limitations GPS or other accelerometers could not be add on the system. The controller obtained the relative positions and velocities from the cameras only. Two methods of control are studied – one using a series of mode-based, feedback lin- earizing controllers and the other using a back-stepping control law. The helicopter was restricted with a tether to vertical, yaw motions and limited x and y translations. Simu- lations performed on MATLAB-Simulink show the ability of the controller to perform output tracking control even when there are errors on state estimates. Fig 2.13: Quadrotor tracking with a camera
  • 40. Chapter 2 LITERATURE REVIEW 40 2.2.4 Middle East Technical University Three orthogonal piezoelectric gyro used in the system designed in Middle East Technical University (Figure 2.14) to control the attitude of the quadrotor. The attitude controlled with a Linear Quadratic Regulator and PD controller. Frame consists of 45 cm rectangular aluminum profiles. Fig 2.14: Quadrotor designed in Middle East Technical University, Turkey 2.2.5 Australian National University The X4-Flyer developed in ANU consists of a HC-12 a single board computer, developed at QCAT that was used as the signal conditioning system. This card uses two HC-12 processors and outputs PWM signals that control the speed drivers directly, inputs PWM signals from an R700 JR Slimline RC receiver allowing direct plot input from a JP 3810 radio transmitter and has two separate RS232 serial channels, the first used to interface with the inertial measurement unit (IMU) and second used as an asynchronous data linked to the ground based computer. As an IMU the most suitable unit considered was the EiMU embedded inertial measurement unit developed by the robotics group in QCAT, CSIRO weighs 50-100g. Crossbow DMU-6 is also used in the prototype. It weighs 475g.
  • 41. 41 The rotor used is an 11’’ diameter APC-C2 2.9:1 gear system included 6’’ per revolution pitch with a maximum trust of 700grams. An the motors are Johnson 683 500 series mo- tors The speed controller used is MSC30 B with a weight 26g rated 30A at 12V The pilot augmentation control system is used. A double lead compensator is used for the inner loop. The final setup is shown in Figure 2.15. Fig 2.15: The X4-Flyer developed in FEIT, ANU 2.2.6 University of British Columbia Vancouver, BC, Canada Setup developed in Department of Electrical and Computer Engineering University of British Columbia Vancouver, BC, Canada. This work focused on the nonlinear modelling of a quad rotor UAV. An experimental system including a flying mill, a DSP system, a programmed microprocessor and a wireless transmitter have been used to test the flight controller. Based on the nonlinear model, an H∞ loop shaping controller is designed for stabilization, speed, throttle and yaw control. A microprocessor, PIC16F877, is pro- grammed to transfer the control data to a pulse width modulated signal in order to reduce significant CPU load which otherwise would have been associated with the DS1102. This signal is further used to control the four rotors of the Draganflyer III via a 4 channel Fu- taba radio transmitter working in training mode.
  • 42. Chapter 2 LITERATURE REVIEW 42 In order to carry out flight control experiments, an experimental rig including a custom designed flying mill, a personal computer, dSPACE DSP board, a microprocessor pulse modulator, a radio transmitter and the Draganflyer III was built. A picture of the flying mill is shown in Figure 2.16. The steel base and carbon fiber boom limit the flight route of the UAV Draganflyer III to a half sphericity of 1 meter radius. Based on the nonlinear model, an H∞ loop shaping controller is designed for stabiliza- tion, speed, throttle and yaw control.Aconstraint model based predictive control (MBPC) controller is implemented for longitudinal and lateral trajectory control. Fig 2.16: Setup developed in Department of Electrical and Computer Engineering University of British Columbia Vancouver, BC, Canada
  • 43. 43 2.2.7 Cornell University The Autonomous Flying Vehicle (AFV) project at Cornell University has been an on- going attempt to produce a reliable autonomous hovering vehicle. In the thrust system MaxCim motors were used. The final vehicle weighed approximately 6.22 kg. Initially an Extended Kalman Filter was designed to handle the estimation of both the state and the six IMU sensor bias parameters. This filter found to be cumbersome to implement, due to extremely large and complex Jacobian terms and instead a square root implementation of a Sigma Point Filter (SRSPF) was considered. The final picture of the quadrotor is shown in Figure 2.17. Fig 2.17: Quadrotor designed in Cornell University 2.2.8 Swiss Federal Institute of Technology In the study done at Swiss Federal Institude of Technology the mechanical design, dy- namic modelling, sensing, and control of an indoor VTOL autonomous robot OS43 is presented. The 3 DOF are locked.From a PC and through a standard RS232 port, orders were sent to the test bench. The RS232 to I2C module translates the serial signals to the I2C bus motor modules. These modules integers a P.I.D regulator on a PIC16F876 micro- controller and are capable of open or closed loop operation in position, speed or torque
  • 44. Chapter 2 LITERATURE REVIEW 44 control. The MT9-B8 IMU9 estimates with a Kalman filter the 3D orientation data and give the calibrated data of acceleration and angular velocity. It weights about 33g and communicates at 115kbps. The captured motion from the 3D universal joint decoded to extract absolute orientation information, by the help of the micro optical encoders in each axis. The cross is made with carbon rods thus vehicle, the mass of which is around 240 grams, is light weight. The OS4 test bench has four propulsion groups, each composed of a 29g motor including magnetic encoders, a 6g-gear box and a 6g propeller. Before implementation on the real system, several simulations had been performed on Matlab. The controller’s task was to stabilize the height while compensating the initial error on the roll, pitch and yaw angles. The real system suffered from undesired but unavoidable delays and actuator saturation. The delays were reported to be mainly due to RS232 communications and the actuator time constant. To emulate these lacks, two Simulink discrete-step delay blocks had been introduced in the feedback loop and on the actuators. Saturation level depends on the chosen actuators. The experimental setup for this study is shown in the Figure 10. Fig 2.18: Quadrotor designed in Swiss Federal Institude of Technology
  • 45. 45 2.2.9 University of Technology in Compiegne, France Quadrotor system, manufactured by the Draganfly Innovations Inc., was used as the test base. The four control signals were transmitted by a Futaba Skysport 4 radio. The radio and the PC (INTEL Pentium 3) were connected using data acquisition cards (ADVAN- TECH PCL-818HG and PCL-726). The connection in the radio is directly made to the joystick potentiometers for the collective, yaw, pitch and roll controls. In order to sim- plify the tuning of the controller and for flight security reasons, several switches were in- troduced in the PC-radio interface so that each control input can operate either in manual mode or in automatic control. Therefore the control inputs that are handled manually were selected by the pilot while the other control inputs are provided by the computer. The Polhemus was connected via RS232 to the PC. This type of sensor was reported to be very sensitive to electromagnetic noise and it was install as far as possible from the electric motors and their drivers. The Draganfly III has three onboard gyros that helped the mini-rotorcraft’s stabilization. The dynamic model of the four rotor rotorcraft was obtained via a Lagrange approach. And the proposed controller was based on Lyapunov analysis using a nested saturation algorithm. The picture of the setup when it was hover- ing is shown in Figure 11. Fig 2.19: Quadrotor designed in University of Technology in Compiegne, France
  • 46. Chapter 2 LITERATURE REVIEW 46 2.2.10 Stanford University The name of the project that is worked on in stanford university is called STARMAC. STARMAC consists of four X4-flyer rotorcraft that can autonomously track a given way- point trajectory. This trajectory generated by novel trajectory planning algorithms for multi agent systems. STARMAC project aims a system fully capable of reliable autono- mous waypoint tracking, making it a useful testbed for higher level algorithms addressing multiple-vehicle coordination. The base system is the off-the-shelf four-rotor helicopter called the DraganFlyer III, which can lift approximately 113,40 grams of payload and fly for about ten minutes at full throttle. The open-loop system is unstable and has a natural frequency of 60 Hz, mak- ing it almost impossible for humans to fly. An existing onboard controller slows down the system dynamics to about 5 Hz and adds damping, making it pilotable by humans. It tracks commands for the three angular rates and thrust. An upgrade to Lithium-polymer batteries has increased both payload and flight duration, and has greatly enchanced the abilities of the system. For attitude measurement, an off-the-shelf 3-D motion sensor developed by Microstrain, the 3DM-G was used. This all in one IMU provides gyro stabilized attitude state infor- mation at a remarkable 50 Hz. For position and velocity measurement, Trimble Lassen LP GPS receiver was used. To improve altitude information a downward-pointing sonic ranger (Sodar) by Acroname were used, especially for critical tasks such as take off and landing. The Sodar has a sampling rate of 10 Hz, a range of 6 feet, and an accuracy of a few centimeters, while the GPS computes positions at 1 Hz, and has a differential accu- racy of about 0.5 m in the horizontal direction and 1 m in the vertical. To obtain such ac- curacies, DGPS planned be implemented by setting up a ground station that both receives GPS signals and broadcasts differential correction information to the flyers.
  • 47. 47 All of the onboard sensing is coordinated through two Microchip 40 MHz PIC micro- controllers programmed in C. Attitude stabilization were performed on board at 50 Hz, and any information was relayed upon request to a central base station on the ground. Communication is via a Bluetooth Class II device that has a range of over 150 ft. The device operates in the 2.4 GHz frequency range, and incorporates band hopping, error correction and automatic retransmission. It is designed as a serial cable replacement and therefore operates at a maximum bandwidth of 115.2 kbps. The communication scheme incorporates polling and sequential transmissions, so that all flyers and the ground station simultaneously operate on the same communication link. Therefore, the bandwidth of 115.2 kbps is divided among all flyers. The base station on the ground performs differential GPS and waypoint tracking tasks for all four flyers, and sends commanded attitude values to the flyers for position con- trol. Manual flight is performed via standard joystick input to the ground station laptop. Waypoint control of the flyers was performed using Labview on the groundstation due to its ease of use and on the fly modification ability. Control loops have been implemented using simple PD controllers. The system while hovering is shown in Figure 2.20. Fig 2.20: Quadrotor designed in Stanford University
  • 48. Chapter 2 LITERATURE REVIEW 48 2.2.11 Australian National University, Canberra, Australia Current work on the X-4 Flyer aims to solve two problems: thrust and stability. X4 Flyer has weight of 2 kg with a length of 70 cm an 11 inch diameter rotor. The electronics are substantially the same as the Mark I, although a lighter sensor unit has replaced the origi- nal Crossbow IMU. The control board and ‘Eimu’ IMU were built by the CSIRO ICT Centre. The control board is a dual HC-12 microprocessor card with digital I/O. The Eimu is a full six-axis IMU with magnetometer. It is operated in vertical gyro mode to obtain inertial frame reference angles. There is room inside the frame for mounting the Eimu as close to the centre of gravity as possible. Unlike the Mark I, the Mark II incorporates simple onboard proportional-integral deriva- tive control. The previous iteration used a slow, off-board control system connected to the flyer by a tether. It is anticipated that the convenient aerodynamics of the X-4 had made sophisticated control unnecessary. In conjunction with onboard power, this allowed the flyer to be entirely self-contained. MARK II shown in figure 2.21.[5] Fig 2.21: X-4 Flyer Mark II
  • 49. 49
  • 52. Chapter 3 skykopter design 52 3 Skykopter Design The most important target of this particular design process is to arrive at the correct set of requirements for the aircraft, which are often summarized in a set of specifications. In this section we will define our mission to build a quadrotor prototype suitable for indoor& outdoor flight, as well justify the decisions and equipment chosen for achieving this pur- pose. The specifications for our quadrotor prototype are: • Overall mass not superior to 1 kg. When it comes to quadrotors, the heavier they get, the more expensive they are. Many quadrotors used for research do not exceed this mass. So, aiming for maximum mass of 1 kg seems to be a suitable target; • Flight autonomy between 10 and 20 minutes. There is no point in using the quadrotor for 2 minutes and then wait a couple of hours to recharge the batter- ies, wasting precious time; • Ability to transmit live telemetry data and receive movement orders from a ground station wirelessly, therefore avoiding the use of cables which could be- come entangled in the aircraft and cause an accident; • The quadrotor should not fly very far from ground station so there is no need of long range telemetry hardware and the extra power requirements associated with long range transmissions.
  • 53. 53 3.1 System Description The Skykopter helicopter used in this project is the Skykopter. It differs from traditional helicopters regarding the design as it has 4 horizontal rotors and no vertical rotors. Tra- ditional helicopters can adjust the angle of the rotor blades as well in order to control the helicopter, but on the Skykopter the blades have a static angle. The only variable that can be adjusted in flight is the rotational speed of the rotors. The reason for this is to aid the development of a model and controllers for the Skykopter, without the risk of flying into the wall. This will of course give reduce the movements of the Skykopter, but it will make it easier to test if the input-output relationship is as expected. 3.2 Components Quad-rotor crafts generally support only a light payload, as they are required to carry the weight of the power supply, a heavy battery, onboard. Thus, weight reduction of all components is essential in order to allow for sufficient lift force. The controls system for such a craft is also complex, as it requires the synchronization of four individual motors.
  • 54. Chapter 3 skykopter design 54 3.2.1 Central Plate Fig 3.1: Central plate The central plate carries all of the electronics, sensors, and battery. It sits lower than the four motors in order to bring the center of gravity downwards for increased stability. We manufactured it using a rapid prototyping machine considering our design for the plate, the rapid prototyping machine was ideal because of its ability to produce relatively com- plex details, for example the angled holes which allow for the central plate to sit lower than the surrounding motors. The thermoplastic polymer used in rapid prototyping has good strength to weight ratio. 3.2.2 Arms Fig 3.2: Arms The arms of our quad-rotor design needed to be light and strong enough to withstand the stress and strain caused by the weight of the motors and the central plate at their opposite ends. Carbon fiber was deemed the best choice because of its weight to strength ratio. The thickness of the tube was chosen to be the smallest possible to lower its weight. The
  • 55. 55 length of each arm (10”) was chosen based on the propellers. The propellers used are 10” long each so we had to allow enough room for them to spin without encountering turbu- lence from one another. Since such a phenomenon would be quite complex to analyze, we simply distanced the motors far enough apart to avoid the possibility of turbulence interference among rotors. [7] 3.2.3 Motors Fig 3.3: Brushless Motor Brushless motors called out runners have become very popular with radio controlled airplane hobbyists because of their efficiency, power, longevity and light weight in comparison to traditional brushed motors. [8] The 12 pole “outrunner” type design creates significant torque and can thus drive direct props without the need for a gearbox. [9] Outrunner are used to implement the proposed system Brushless DC motor In aircrafts it’s all about power to weight ratio. Outrunner brushless motors produce more torque output than conventional brushless motor. A motor can produce power up to 90 W when it only weights 48g. The motors are cobalt, brushed, DC motors rated for 12 V, 15 A. The DC, brushed motor configuration was desired for ease to control via PWM (Pulse-
  • 56. Chapter 3 skykopter design 56 width modulation). Each one of these brushless motors uses a special electronic speed control (ESC). The ESC’s accept a 50 Hz PWM signal with a 1 millisecond duty cycle corresponding to stopped and a 2 millisecond duty cycle corresponding to full power. The cobalt motors use strong rare earth magnets and provide the best power to weight ratio of the hobby motors available for model aircraft. We were limited to these hobby motors by our design budget. As a result, the rest of our structural design revolves around the selection of these motors and the allowable weight of the craft based on the lift provided by these motors. 3.2.4 Propellers Fig 3.4: Propellers The propellers are 10” from tip to tip. Two are of the tractor style, for clockwise rotation, and the other two are of the pusher style, for counterclockwise rotation. For our design, a propeller with a shallow angle of attack was necessary as it provided the vertical lift for stable hovering. The propellers we used were steeper than the ideal design because of lim- ited availability of propellers that are produced in both the tractor and pusher styles. [7] The lower pitch angle lead to less torque per turn thus more spin speed.Using the selected motors and propellers the quadrotor is capable of maximum theoretical liftoff weight of 2.8Kg. [9]
  • 57. 57 3.2.5 Flight Control Fig 3.5: Flight controller The flight control board of the model after soldering the gyros, accelerometers, pressure sensors. The main board of the quadrotor has 3 single axis MEMS based gyros for esti- mating the tilting degrees for each axis (Euler angles; Phi, Theta, Epsi) while the 3-axis MEMS based accelerometer is to estimate the linear acceleration of the body and hence we can find the velocity and the displacement. Also there is a pressure sensor for estimat- ing the altitude of the quadrotor in flight and the main part of the board is the Atmel AT mega microcontroller which will role all of that to achieve a stationary stable flight, one of sensors is Linear Accelerometer which is a low-power three axes linear accelerometer that includes a sensing element and an IC interface able to take the information from the sensing element and to provide an analog signal to the external world. The chip is powered by 3.3V and outputs a voltage proportional to acceleration. The ac- celerometers can be used to determine differential position through integration. Addition- ally, the sensor measures acceleration due to gravity and can be thus used to determine tilt or orientation of a static object, and Gyroscope Sensor is to measure the angular velocity. The output signal is a voltage proportional to the rate of rotation about the axis normal to the top surface of the chip. Regular sampling of this voltage signal was required to detect the rate of rotation of the module. By 3 sensors of this type, the angular velocities of the X, Y and Z axis determined. These data are used to stabilize the Quad-rotor.
  • 58. Chapter 3 skykopter design 58 3.2.6 Brushless control OR ESC (Electrical Speed Control) Fig 3.6: Electrical speed control The BL-Ctrl board is a sensor less driver for brushless DC current motors. The ESC (electrical speed control) board is a sensor less driver for brushless DC current motors. They are especially designed for use in quadrotor, where quick set-changes are necessary. The control of brushless motors is triphasic in groups of PWM pulses. The PWM frequency determines the height of the phase-voltage. The ESC’s accept a 50 Hz PWM signal with a 1 millisecond duty cycle corresponding to stopped and a 2 millisec- ond duty cycle corresponding to full power. 3.2.7 Battery Fig 3.7: Battery The battery was selected on the basis of power requirements for the selected motor/gear- box combination. We opted for a battery of the lithium-polymer variety, despite the fact that it was considerably more expensive than other batteries providing the same power, because this battery provided the best power-to-weight ratio. Our battery choice was a
  • 59. 59 2200mah (Milliamp Hour) 12.0V 12C Li-polymer battery. (Note: Because we did not have enough time to integrate the circuitry of the controls system on-board, and thus performed only tethered flight, we did not ultimately pur- chase the battery.)[7] 3.2.8 Sender (MX-16s Transmitter 35/35B MHz) Fig 3.8: MX-16s transmitter Microcomputer control system for 8 control functions (channels) High-Technology micro-computer system with new high-speed single-chip micro- computer, flash memory and 10-bit A / D Converter With advanced technology opti- mized computer radio control system with 12 model memories. High operational safety through modern Computer system. Easy programming through simplified programming technique with rockers and buttons. High-contrast graphic screen provides an efficient control of the parameters, operating modes, timers and battery voltage. Modern hard- ware and integrated Synthesizer system for channel selection, with security against ac- cidental start-up. Configuration and programming based on proven concepts of MC-19 MC-24.8 and control functions with a simplified method of assigning controls for auxil- iary functions such as external switches and proportional donors provide high efficiency. Free assignment of all switches to switched functions simply by operating the desired Switches. 12 model memories.
  • 60. Chapter 3 skykopter design 60 3.2.9 Receiver (DSL-4TOP) Fig 3.9: DSL-4top This receiver is a receiver for remote controlled aerial vehicles like model helicopters. Its output signal is a sort of pulse width modulated. A single-receiver with the highest range and selectivity Special version with multi-signal (sum-signal) output The DSL 4top is a small 4(9)-channel micro-receiver of the company ACT. 3.2.10 Bluetooth Fig 3.10: Bluetooth module Is an open wireless protocol for exchanging data over short distances from fixed and mobile devices, creating personal area networks (PANs). It was originally conceived as a wireless alternative to RS232 (Recommended Standard 232) is a standard for serial bi- nary single-ended data and control signals connecting data cables.
  • 61. 61 A Bluetooth module is mounted on the flying machine and it is attached to the main board. A wireless connection is created between the flying machine and ground based computer (must have Bluetooth card). Another advantage granted by Bluetooth connec- tion it provides secured data line between the ground station and the mobile device. The F2M03GXA is an embedded Bluetooth 2.0 EDR with excellent radio transmission char- acteristics, highly efficient, Omni directional on-board antenna and serial port profile (SPP). On the module is running by default, the extremely reliable and easy to use Wire- less UART firmware that implements the Bluetooth Serial Port Profile (SPP). Any infor- mation that is received at the serial interface is transmitted transparently via Bluetooth to the connected remote device. The ‘pairing’ of two modules is possible. In that case a greater range than with a simple Bluetooth dongle can be achieved (1000m).
  • 62. Chapter 3 skykopter design 62 3.3 Mechanical design 3.3.1 Frame Fig 3.11: frame The Quadrotor consists of a base in the center and 4 arms equally spaced in a shape of a cross. The motors are mounted on the end of the arms and connected to the rotors. The base of the Quadrotor is the joint between the 4 arms and contains a battery and main electrical control component. Four DC Motors mounted at the four corners of a square, this configuration obviates the need for a tail rotor to produce the anti-torque required to counter the main rotor torque. One pair of diagonally opposite rotors rotates in a clockwise direction and another pair in counter-clockwise direction, nullifying the torque on the MAV (Micro Air Vehicle) due to the former pair.
  • 63. 63 3.3.2 Brushless motors. Fig 3.12: Brushless motor The direct current (DC) motor is one of the first machines devised to convert electrical energy to mechanical power. Its origin can be traced to machines conceived and tested by Michael Faraday, the experimenter who formulated the fundamental concepts of electro- magnetism. These concepts basically state that if a conductor, or wire, carrying current is placed in a magnetic field, a force will act upon it. The magnitude of this force is a function of strength of the magnetic field, the amount of current passing through the conductor and the orientation of the magnet and conductor. The direction in which this force will act is dependent on the direction of current and direction of the magnetic field. As the name implies, BLDC motors do not use brushes for commutation; instead, they are electronically commutated. BLDC motors have many advantages over brushed DC motors and induction motors. A few of these are: • Better speed versus torque properties • High dynamic response • High efficiency • Long operating life • Noiseless operation • Higher speed ranges
  • 64. Chapter 3 skykopter design 64 BLDC motors come in single-phase, 2-phase and 3-phase configurations. Corresponding to its type, the stator has the same number of windings. Out of these, 3-phase motors are the most popular and widely used. This application note focuses on 3-phase motors. Fig 3.13: Three phase configuration The stator of a BLDC motor consists of stacked steel laminations with windings placed in the slots that are axially cut along the inner periphery. Traditionally, the stator resembles that of an induction motor; however, the windings are distributed in a different manner. Most BLDC motors have three stator windings con- nected in star fashion. Each of these windings are constructed with numerous coils in- terconnected to form a winding. One or more coils are placed in the slots and they are interconnected to make a winding. Each of these windings is distributed over the stator periphery to form an even numbers of poles.
  • 65. 65 3.3.3 Propellers Fig 3.13: Pusher and puller styles of propellers The propellers are 10 inch from tip to tip. Two are of the tractor style, for clock wise ro- tation, and the other two are of the pusher style, for counter clock wise rotation. For our design, a propeller with a shallow angle of attack was necessary as it provided the vertical lift for stable hovering. The propellers we used were steeper than the ideal design because of limited availability of propellers that are produced in both the tractor and pusher styles.
  • 66. Chapter 3 skykopter design 66 3.4 Electrical Design Skykopter is piloted with the Spectrum MX-16s which has 8 programmable channels for signal mixing and processing. Remote pilot input is transmitted to a receiver onboard the aircraft which is wired to a separate speed controlled unit for each of the four rotors. In general, a speed controller functions by accepting an electrical signal from the receiver to regulate the voltage delivered from the battery to the motor. 3.4.1 Electronic Speed Control (ESC) The ESC board is a sensorless driver for brushless DC current motors. They are especially designed for use in quadrotor, where quick set-changes are necessary. The control of brushlss motors is tri-phasic in groups of PWM pulses. The PWM frequency determines the hight of the phase-voltage (actually the arithmic mean of the voltage). The phase voltage on the motor (as group of PWM pulses) determines the rotations per minute (RPMs) of the motor: a motor generates a countercurrent when turning (like a generator). the resultant RMPs will be the balance between current, coutercurrent and power output. At any time 2 of the 6 FETs are operational to power the motor. The time of commutation (the time at which the 2 FETs must be switched off and the next two swit- ched on to go to the next phase), is determined by voltage measurement (actually com- parison) of the unpowered phase of the three motor wires. To do this the different analog comparators of the Atmega8 are used. 3.4.2 Controller: Atmega644 8-bit microcontroller TheATmega644 is a low-power CMOS 8-bit microcontroller based on theAVR enhanced RISC architecture. By executing powerful instructions in a single clock cycle, the ATme- ga644 achieves throughputs approaching 1 MIPS per MHz allowing the system designed
  • 67. 67 to optimize power consumption versus processing speed. One of the microcontroller’s functions and the one that it provides in the control part is to take operator inputs (thrust, pitch, roll and yaw) form the receiver (PPM: Pulse Posi- tion Modulation Signal) and process it then transferred to the ESCs (PWM: Pulse Width Modulation Signal) 3.4.3 Implementation of the interfacing circuits The receiver is connected to the Atmega644 through pin PD6, the transmitters signal is obtained by the receiver then transferred to the microcontroller as a PPM signal. After that the signal is processed and compared with data coming from the accelerometers and gyroscopes (will be explained later in the chapter). Four Brushless ESCs are connected in parallel configuration to pins PC1 and PC0 used for SDA (Serial DAta) and SCL (Serial CLock) respectively. Data is transferred to the four ESCs. Each ESC is identified indi- vidually using 3 nodes by connecting or disconnecting (soldering or not soldering) these points in four different configuration. The Brushless ESCs controls the corresponding ro- tors speed using PWM signal. The integrated system acts as a closed-loop system to give the stability desired and maximum control on the Quad-rotor. Fig3.15: Control Block Diagram Receiver: Operator inputs (Thrust, tilt and direction) Microcontrollers Sensors: Feed back (Thrust, roll, yaw and pitch) ESCs Motors
  • 68. Chapter 3 skykopter design 68 3.4.4 Pulse Width Modulation (PWM) Pulse Width Modulation, or PWM, is a technique for getting analog results with digital means. It may be used as an efficient light dimmer or DC motor speed controller. The circuit described here is for a general purpose device that can control DC devices which draw up to a few amps of current. The circuit may be used in either 12 or 24 Volt systems with only a few minor wiring changes. This device has been used to control the brightness of an automotive tail lamp and as a motor speed control for small DC fans of the type used in computer power supplies. A PWM circuit works by making a square wave with a variable on-to-off ratio, the aver- age on time may be varied from 0 to 100 percent. In this manner, a variable amount of power is transferred to the load. The main advantage of a PWM circuit over a resistive power controller is the efficiency, at a 50% level, the PWM will use about 50% of full power, almost all of which is transferred to the load, a resistive controller at 50% load power would consume about 71% of full power, 50% of the power goes to the load and the other 21% is wasted heating the series resistor. Load efficiency is almost always a critical factor in solar powered and other alternative energy systems. One additional advantage of pulse width modulation is that the pulses reach the full supply voltage and will produce more torque in a motor by being able to overcome the internal motor resistances more easily. Finally, in a PWM circuit, common small potentiometers may be used to control a wide variety of loads whereas large and expensive high power variable resistors are needed for resistive controllers. The main Disadvantages of PWM circuits are the added complexity and the possibility of generating radio frequency interference (RFI).
  • 69. 69 Fig 3.14: PWM The main use of PWM is in power electronics, for example for motor control. When gen- erating analogue signals, the bad thing is that PWM resolution goes rapidly down with required signal bandwidth. For example, 100 kHz PWM period could be sufficient for generating hi quality audio signal without the need of complicated analogue output filters. With 20 MHz base frequency you get 7.64-bit resolution, with 100 MHz you get to 10 bits. This is good for toys, but a hi-fi audio signal needs 16-bit resolution.
  • 70. Chapter 3 skykopter design 70 3.5 Basic Control of the Skykopter The Skykopter is controlled by alternating the rotational velocity of the rotors. The front and back rotors are rotating counter clockwise and the rotors to the left and right are ro- tating clockwise. A main board is part of the electronics on the Skykopter. It handles the incoming control signals in the form of throttle, yaw, pitch and roll and translates them into motor control signals. Yaw is also known as the heading. The main board makes it possible to control the Skykopter like a traditional helicopter. 3.5.1 Throttle When throttle is increased the revolution on the four motors increases simultaneously. The rotors will then create a lift and the Skykopter will increase in altitude. In figure 3.16 the bars indicate that the rotational speed is above average. The Skykopter is seen from above and the black bar is the front. Fig 3.16: Illustration of the Skykopter with increased throttle which results in increased altitude
  • 71. 71 3.5.2 Yaw When the heading is changed the overall lift is maintained, although the speed of the ro- tors changes. To make the Skykopter rotate clockwise the speed of the left and right ro- tors are decreased and the speed on the front- and back rotors are increased. This way the Skykopter maintains the same lift force. Furthermore, the Skykopter increases the drag force from the two rotors, with the highest velocity, which makes the Skykopter rotate. An illustration of the Skykopter turning clockwise can be seen in figure 3.17. To make it turn counter clockwise the opposite is obviously applied. Fig 3.17: Illustration of the change of rotor speed needed to turn the Skykopter clockwise.
  • 72. Chapter 3 skykopter design 72 3.5.3 Pitch Pitch is used in order to make the Skykopter move forward and backwards. Moving for- ward is a result of tilting the Skykopter forward, where the velocity of the front rotor is decreased, and the back rotor is increased. During this movement the total lift needs to be maintained, keeping the velocity of the side rotors unchanged. Figure 3.18 illustrates the Skykopter pitching backwards. The opposite happens when tilting the Skykopter back- wards. Fig 3.17: Illustration of the changing rotor speed, needed to tilt the Skykopter backwards.
  • 73. 73 3.5.4 Roll Roll is used to make the Skykopter move side wards. Roll is performed like pitch, except that the velocity of the side rotors are changed, instead of the front- and back rotors. Fig 3.19: Illustration of change of rotor speed needed to make the Skykopter roll to the left.
  • 74. Chapter 3 skykopter design 74 3.6 System Blocks The overall system has been described, and the system is now divided into blocks, in or- der to make an analysis. A block diagram of the system blocks is illustrated in figure 2.8. Fig 3.20: Blocks in the system. 3.6.1 R/C Transmitter and Receiver The first block is the R/C transmitter that sends 4 control signals (throttle, yaw, pitch and roll). This is sent to the R/C receiver that passes the signals to the Skykopter main board. The data is sent as pulse signals. 3.6.2 Main board (flight control) Increasing or decreasing throttle or yaw signal will affect all the motors, while pitch and roll only affects 2 of the motors. The input received by the Skykopter is the only informa- tion about throttle, pitch, yaw and roll. The main board translates this information into motor control signals, which is further described in appendix A.
  • 75. 75
  • 77. Chapter 4 modeling, simulation, Testing and validation
  • 78. Chapter 4 MODELING AND SIMULATION 78 4 modeling and simulation 4.1 Dynamic Simulation Dynamic attitude model was derived from Newton’s Laws. The gyroscopic Procession of each rotor cancels out due to the counter-rotating pairs, which removes any coupling between the pitch and roll dynamics. Due to the low rotor inertia relative to the craft’s ro- tational inertia, the response of the electric motor was faster than the attitude dynamics, so the motor response was neglected in this model. The total collective thrust FT is the sum of all four rotor forces. (Subscripts are T = Total, F = Front, B = Back, L = Left, R = Right) FFFFF RLBFT +++= [10] (4.1) This collective thrust is nominally equal to the gravitational force when hovering; how- ever, it can be varied by the pilot with the throttle input up to a maximum of Fg ×2 , due to the 2:1 thrust/weight ratio. When quad rotor is in a stationary hover, FT equals the weight force of the entire aircraft due to gravity. Aerodynamics, forces and moments Fig ‎4.1: Plus configuration for flight control
  • 79. 79 4.1.1 Throttle In figure 4.2 a step applied to the throttle on both the non-linear and the linear model, is shown. Both models react like expected since the z position is increasing. It is not expect- ed that the throttle step should change the Euler angles, which can be verified in figure 4.2. The two models keep the same angle values. Fig 4.2: Position values when applying a step to the throttle input on the linear and non-linear model. And Euler angle values when applying a step to the throttle input on the linear and non-linear model.
  • 80. Chapter 4 MODELING AND SIMULATION 80 4.1.2 Yaw The step applied to the yaw input is shown in figure 4.3 and makes the models change in z values, which not affect the x and y position of the models. The positive step on the yaw input results in a negative Euler angle on the z axis. The linear model reacts like the non-linear model, but is more aggressive. The linear model is sufficient enough, and will be used in designing the state-space controller. Fig 4.3: Plot of the positions affected by a step on pitch input. And plot of the Euler angle affected by a step on the yaw input. 4.1.2.1 Yaw dynamics A quad rotor has two sets of counter-rotating propellers; therefore, the net yaw moment generated from aerodynamic drag is cancelled out in neutral flight. This eliminates the need for a tail rotor that normally wastes 12% of the power in a conventional helicopter. Furthermore, a yaw moment is induced on a quad rotor by proportionally varying the speeds of the counter-rotating pairs. The thrust variationVψ is given by: ).max,42( max max Torquesaturationmotoravoidtoork kV ==≤ τ τ ψ [10] (4.2)
  • 81. 81 From ψτψ ×= Iz , where Iz is the mass moment of inertia and yaw acceleration is )( angleYawwhere Iz ==Ψ ψ τψ [10] (4.3) Yaw moment is the sum of all rotor torques (CW = Clockwise, CCW = counterclockwise) )()( ττττττττψ BFRLCCWCWr +−+=−==∑ [10] (4.4) The magnitudes of thrust forces are set so that FF RL = are both larger than FF BF = . The increased drag of the motors with higher thrust will create a net reaction moment that will rotate the body in one yaw direction. Similarly, the body can be rotated in the opposite yaw direction by reversing the relative magnitudes of the above pairs of thrust forces. Note that during yaw movement of the quad rotor 0≠τψ (net torque on the body), i.e. the sum of reaction moments is non-zero. When the quad rotor body is not rising or dropping in altitude, the sum of all thrusts equals the weight force due to gravity. The torque on each rotor, caused by aerodynamic drag, is proportional to the thrust by a scalar constant kτ Therefore, Equation (4.4) becomes: VkVkVkVkVk ψτψτψτψτψτψτ 4)()( =−−−+= [10] (4.5) The z-axis “Moment Of Inertia” (MOI) of the quad rotor is the sum of all point mass in- ertias about the z-axis (assuming battery and controller inertia is negligible due to their masses being located predominantly at the “Centre of Gravity”, or COG). )&(4 2 massarmmotoraiswheremlmII mmmz ==∑ [10] (4.6) Therefore, substituting Equations (4.5) and (4.6) into (4.3), gives the equation of motion for yaw acceleration: lm Vk lm Vk I mmz 22 4 4 ψτψτψτψ === [10] (4.7)
  • 82. Chapter 4 MODELING AND SIMULATION 82 4.1.3 Roll When applying a step on the roll input the models results in positive x direction values, shown in figure 4.4. The linear model is decreasing faster than the non-linear, which is caused by the gyroscope effect on the non-linear model. The step on the roll input in fig- ure 4.4 have a positive Euler angle effect around the x axis. In The figures it can be seen that the models act as expected. Fig 4.4: Position values when applying a step to the roll input on the linear and non-linear model. And Euler angle values when applying a step on roll input on the linear and non-linear model.
  • 83. 83 4.1.4 Pitch The step applied on the pitch input is shown in figure 4.5 and makes both models move in the positive x direction. Both models react as expected, but as on roll step, the non-linear is less aggressive due to the gyroscope effect. The angles shown in figure 4.5 for the roll step, results in positive y values. Fig 4.5: Position values when adding a pitch input step on the linear and non-linear model. And Euler angle values when adding a pitch input step on the linear and non-linear model.
  • 84. Chapter 4 MODELING AND SIMULATION 84 4.1.4.1 Pitch and roll dynamics Due to the symmetrical nature of the quad rotor configuration, pitch and roll can be rep- resented by the same model. Figure 4-6 illustrates the thrust variations required to induce a moment about the y-axis for rolling. The yaw deviation limit is thus ),42( max max , ForceMaximumsaturationmotoravoidtoork k F F V ==≤θφ [11] (4.8) The equation of motion for this pitching or rolling moment is derived from the sum of moments about the y-axis: ∑ ×= θτθ  Iy [11] (4.9) The thrust deviation for one motor can be calculated as 2/)( FFV FB −=θ [11] (4.10) Therefore the sum of the moments is lV∑ = θθτ 2 [11] (4.11) The y-axis moment of inertia of quad rotor is the sum of the two point mass inertias ∑ == lmII mmy 2 2 [11] (4.12) We now substitute Equations (4.11) and (4.12) into (4.9) to find pitch acceleration. l l m V lm V mm θθ θ == 2 2 2  [11] (4.13) Due to symmetry of the quad rotor body, this also represents pitch dynamics. The dy- namic equations discussed so far have treated the quad rotor as a flying “+” structure. The above equations for the Quadrotor model were taken from “Advances in Unmanned Aerial Vehicles” by (S.Bouabdallah, R. Siegwart), it is an advanced model, hence we thought about simplifying that model.
  • 85. 85 4.2 System of Differential Equations The dynamics of the quad-rotor UAV are determined from a set of equations of motion. The complexity of the equations of motion increases with increased accuracy. The set of equations presented model the motion of the craft based on the amount of lift delivered by each individual motor without taking into account the aerodynamics of the craft. Motion is obtained by varying the amount of lift each motor provides. The amount of lift each motor provides is controlled by the amount of power delivered to each motor. Depending on the motor, the relationship can be linear, parabolic, or a combination of various other trigonometric functions. For example, the craft will move in the positive x-direction by reducing the thrust from motor A (by reducing the power) and simultaneously increasing the thrust (by increas- ing the power) from motor C. The thrust from motor B and D must be increased so that the craft maintains constant altitude while moving along the desired path. More complex movements can be achieved by varying the power delivered to all four motors. The equa- tions of motion that govern the dynamics of the craft are listed below Linear Acceleration in the X-axis Direction: m u k • •• Χ⋅−+ =Χ 1 )cossincossin)(sin1( φθψφψ [11] (4.14) Linear Acceleration in the Y-axis Direction: m u k • •• Υ⋅−+ =Υ 2 )sincoscossin)(sin1( φψφθψ [11] (4.15) Linear Acceleration in the Z-axis Direction: g m u k − Ζ⋅− = • •• Ζ 3 coscos)1( φθ [11] (4.16)
  • 86. Chapter 4 MODELING AND SIMULATION 86 Pitching angular Acceleration in the X-axis Direction: Ι       −= ••• y iu k θθ 5 )2( [11] (4.17) Rolling angular Acceleration in the X-axis Direction: Ι       −= ••• x iu k φφ 4 )3( [11] (4.18) Yawing angular Acceleration in the X-axis Direction: Ι       −= ••• z iu k ψψ 6 )4( [11] (4.19) Equations of vertical forces produced by the four rotors: FFFFu 4321 )1( +++= [11] (4.20) FFu 13 )2( −= [11] (4.21) FFu 24 )3( −= [11] (4.22) FFFFu 4321 )4( −+−= [11] (4.23) : Vertical forces produced by four rotors : Moments of inertia directions : Pitch angle, roll angle, yaw angle : Drag coefficients FFFF 4321 ,,, III zyx ,, ψφθ ki ,...3,2,1=
  • 87. 87 4.3 Assembly of the model The battery, camera, electronic circuits will be acting by weight forces only and the figure below shows the assembly of the model and showing the main axes of direction that will be considered the positive direction while studying the Simulink model of the quadrotor. Fig 4.6: Quadrotor Assembly Where i, j, k are the unit vectors associated to those axes.
  • 88. Chapter 4 MODELING AND SIMULATION 88 4.3.1 Mass and inertial properties 4.3.1.1 Mass properties These properties were estimated using the inspection tool of Solid edge, the mass properties of each part is illustrated below. Item )Mass (grams Quantity Total mass battery 223 1 223 Square tube 26 4 104 Central plate 4 2 8 Control circuits 90 1 90 DC motor 43 4 172 ESC 6 4 24 receiver 9 1 9 propeller 7 4 28 Fasteners 30 1 30 Quadrotor 688 Total 688 4.3.1.2 Inertial properties These properties were estimated using the inspection tool of Solid edge to be used in the Simulink model parameters, the inertial properties of each part is illustrated below Axis M.M. Inertia Radii of Gyration X kg.m^2 0.003325 mm 88.19 Y Kg.m^2 0.003263 mm 87.36 Z Kg.m^2 0.006430 mm 122.64
  • 89. 89 4.4 Simulink model 4.4.1 Simplified quad-rotor Simulink model Fig 4.9: Simplified Quad-rotor Simulink Model The simplified model of the quad-rotor consists of three subsystems: • Resultant Force Diagram: The inputs of this block diagram are the resultant forces (F1, F2, F3 and F4 of the motors) and it calculates the sum of forces acting on the quad-rotor due to variation of motors speeds in order to estimate the transitional and rotational motion of the quad-rotor. • Euler Angles: This block calculates Euler angles (Phi, Theta and Epsi) as a result of variation of forces acting on the quad-rotor. • System Dynamics: It contains the equations of the system dynamics and it calculates the position of the quad-rotor at any given time.
  • 90. Chapter 4 MODELING AND SIMULATION 90 This is a cut down model and has a clear simple mathematical modeling. Some of the pa- rameters affecting the system were neglected such as ground effect and residual angular speed of the motors. The simplified quad-rotor model is based on Quad-rotor dynamics equations explained previously. 4.4.2 Simulink of basic control system 4.4.2.1 Backstepping technique In control theory, backstepping is an analysis technique for designing stabilizing controls for a special class of nonlinear dynamical systems. Because of the recursive structure these systems; the designer can start the design process at the known-stable system and “back out” new controllers that progressively stabilize each outer subsystem. The process terminates when the final external control is reached. Hence, this process is known as backstepping [12] and [13]. Fig 4.10: Simulink of basic control system The block diagram in Figure 4.10 explains how the controller of the quad-rotor process data from sensors about the quad-rotor and the surrounding environment and execute commands applied by the operator accompanied by achieving stability of the system to insure safe flight.
  • 91. 91 4.5 Hardware Testing Before attempting to fly the quadrotor we had to perform some tests to insure that these sensors are functional. Bluetooth communication was established between the quadrotor and a PC. Tests were applied by manual moving or rotating the quadrotor. The diagram shown in Figure 4.10 was obtained by manually rotating the quadrotor around the Y axis then around the X axis followed by random movement. The figure shows the response of the angular rotation (pitch and roll) of the physical. Fig 4.10: Gyro Testing
  • 92. Chapter 4 MODELING AND SIMULATION 92 The same procedure used in testing gyros was applied to test the accelerometer. Figure 4.11 shows the response of angular acceleration of the physical model. Fig 4.11: Accelerometer testing
  • 93. 93 In order to validate the Simulink model, a comparison between the simulation model and the physical model is made using a step input for the two systems. Figure shows the out- put from the two systems which indicate an accepted result. Figure 4.12: the comparison between the Simulink model and the hardware
  • 96. Chapter 6 CONCLUSION 96 5 Coclusion and future work 5.1 Conclusion The aim of our project is to have fully functional quad-rotor helicopter capable of hover and directional hover based on operator inputs. It has high stability. Also, it has the ability of hovering at low altitudes, the system has been heavily tailored to indoor flights .It has the ability to roll, pitch and yaw. The estimated mass and inertial properties from the CAD model were very close to the real model. The design and construction of the physical structure was a success both in terms of stiffness and strength of the craft, and in terms of weight reduction. It can carry 1000 grams additional to its weight. It’s capable of monitor- ing the altitude and the pressure through the gyros and pressure sensors. 5.2 Future work It can be considered as future work to optimize the control law by making feedback gain values depend on the command inputs, Display the vehicle position and other data re- ceived through manual radio control and video reception. Provide autonomous flying in caves, tunnels, and urban environments demands. Put 4 distance sensors to provide com- plete vision of its surroundings. Add heat sensors to provide the ability of mind detector. Develop an efficient algorithm for stability analysis and stabilization. Develop a larger quad-rotor platform than is typically used in current robotics research. Analyze flying at- titude dynamics to allow us to tune the mechanical design for the best control sensitivity. Provide the ability to control it through the cell phone.
  • 97. 97
  • 98. References 1. Leishman, J.G., The Breguet-Richet Quad-Rotor Helicopter of 1907. Verti flite. 2001. 2. http://www.nationmaster.com/encyclopedia/Quadrotor. 2009. 3. http://en.wikipedia.org/wiki/Moller_Skycar_M400. 2009. 4. http://en.wikipedia.org/wiki/M200G_Volantor. 2009. 5. Valavanis, K.P., Advanced in Unmanned Aerial Vehicle, state of the art and the road to autonomy. 2007. 6. KIVRAK, A.Ö., DESIGN OF CONTROL SYSTEMS FOR A QUADRO TOR FLIGHT VEHICLE EQUIPPED WITH INERTIAL SENSORS. 2006. 7. Carlo Canetta, J.C., Sevan Mehrabian, Ludguier Montejo, Hendrik Thomp son, Quad-rotor Unmanned Aerial Vehicle. 2007. 8. Domingues, J.M.B., Quadrotor prototype. 2009. 9. Rami AbouSleiman, D.K., Ermal Gjioni, and Hong Chul Yang, Unmanned Aerial Quadrotor System. 2008. 10. KimonP.Valavanis, ed. Advances in Unmanned Aerial Vehicles. INTELLIG ENTSYSTEMS,CONTROL,ANDAUTOMATION: SCIENCEANDENGI NEERING, ed. S.G.Tzafestas, Springer. 11. kemp, Modeling of flight dynamics of a quad rotor helicopter, in School of Engineering, Aerospace sciences, Cranfield University. Master of Science by research 12. Kokotovic, P.V., The joy of feedback: nonlinear and adaptive in Control Systems Magazine. 1992. p. 7–17. 13. Khalil, h.k., Nonlinear Systems 3rd ed. 2002: Prentice Hall