Weitere ähnliche Inhalte
Ähnlich wie Optimized solutionfor image processing through mobile robots work
Ähnlich wie Optimized solutionfor image processing through mobile robots work (20)
Mehr von IAEME Publication
Mehr von IAEME Publication (20)
Kürzlich hochgeladen (20)
Optimized solutionfor image processing through mobile robots work
- 1. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
140
OPTIMIZED SOLUTION FOR IMAGE PROCESSING THROUGH
MOBILE ROBOTS WORKING AS A TEAM WITH DESIGNATED
TEAM MEMBERS AND TEAM LEADER
Kabeer Mohammed,
Dr.Bhaskara Reddy Ph.D(C.S)
Professor
Department of Computer Science and Technology.
Sri Krishnadevaraya University
ABSTRACT
In the field of Artificial intelligence Image processing And Analysis plays a
Major role in Decision making .Now a day’s Mobile robots work as a Team for
specific applications this results in efficiently managed work process..All Image
inputs are compared against backend database and decisions are made. The Whole
team works is specified work area or sometimes spread around the country or globe.
These teams can be part of any Military Installitions, Surigical Team, Surveying Team
or part any teaching Staff. This Research paper is about how to enhance image
processing/Analysis through Mobile robots. This research paper compares Image
Processing by multiple robots working together but not as a team with same mobile
robots working as team with designated team members and team leader. In this
dangerous situations quick Decision making makes the difference between Hit and
Miss this can also results in Day to day tasks performed by Mobile robots Successful
or Failure.
Keywords: Team, Mobile Robots, Image analysis, Global Team.
INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING
& TECHNOLOGY (IJCET)
ISSN 0976 – 6367(Print)
ISSN 0976 – 6375(Online)
Volume 4, Issue 3, May-June (2013), pp. 140-148
© IAEME: www.iaeme.com/ijcet.asp
Journal Impact Factor (2013): 6.1302 (Calculated by GISI)
www.jifactor.com
IJCET
© I A E M E
- 2. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
141
INTRODUCTION
Due to Application of Mobile robots in Diversified Fields theircontrol,
coordination among RobotsPlays a Vital role in Decision making to complete any
assigned task .Now a day’s Mobile Robots are applied in various fields likeMilitary,
Medical, Aeronautics etc and Decisions are taken after processing input which can in
the form of images entering through its Cameras or Audio input, and through many
types of Sensors. This input is compared against already existing knowledge base this
knowledge base some time is centralized and all mobile robots connect it through wire
or wireless in either case Input need to processed very quickly so that Mobile robot
can Reach and execute the Decisions made Quickly and accurately.
RELATED WORK
1)Mobile Robots which work in an assembly line is a manufacturing process (most of
the time called a progressive assembly) in which parts (usually interchangeable parts)
are added to a product in a sequential manner to create a finished product much faster
than with handcrafting-type methods. The Mobile robots which fall under this
category are applied in Automobile industry, Garment Industry, Beverages,
Medical[1][3]
2)Department of Defense (DoD) Robots.[2]
The Mobile robots of above mentioned Progressive Assembly setup are used in
Automobile Industry. The production stops if first stage robots fails or break down
stooping process flow resulting in Production, Wastage of resources [2].these robots
are used in Defense Field.
1. Problem Description
The above mentioned robots which are working in Progressive Assembly setup
.In this type of setup the output of first mobile robot is feed as input to the next one
and so on. The mobile robots are inter-dependable on each other[7][6]. If Due to any
reason if the first Mobile breakdown then all following mobile robots will be useless
which results in Wastage of Time[8], Resources and hit on Production. This same is
applicable to any mobile robot break down effects on downstream mobile robots
Results in more Idle time for Robots[7] .This setup is suitable for Non- critical
application. Where Decision making is critical this approach may not be best suited
[1],[2],[3].
a. Wastage of time.[1][6]
b. Mobile idle time will increase, Resources are not utilized fully [1]
c. Not suitable for critical time sensitive Production lines [1][2][7]
d. Inefficient in Decision making.[1][2]
- 3. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
142
The problem is we have performance degradation using previous methodologies.
The conventional robots use method 1 in which every input is dependable on previous
stage makes the whole process more inter-dependable that any upstream failures will
result in Down Stream breakdown also [1].
Suggested Solution
Our solution provides setting all mobile robots in Team leader and team
members Format. There will be one designated robot which will Team leader who will
maintain and monitor remaining mobile robots. All remaining Mobile robots will
work as Team members. The team lead robot will first look at Total data to processed
and divide the work load into 5 equal parts this division is based on number of
member robots. Team lead robot maintains Starting and Ending Relative record
number of each assigned work load. In This method Member objects are not inter-
dependable and failure of one will not affect other members and also does not
negatively effect on Production or Production time[8]. If any Team member fails the
team leader which is monitoring all members will assign that work to other member
who has completed his part of work this way we are able to utilize resources to full
extent. Decision making capability is done by Team leader robot. Our method has
many benefits such as:
1. Increased decision making capability.
2. Full utilization of Resources.
3. Increase in number of Images Processed.
4. Increase in Error handling Capability.
5. Increase in Team co-ordination and better results.
The below flow chart explains the Bhaskara & Abuhamza-Shereen Method
Whatever is suggested Is explained in detailed flow how Input images are processed
and how Mobile robots
Work in Team lead and Team member format.You can also See How Team
leader TM1 Mobile Robot Monitors and Maintains other Member Robots.
- 4. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
143
Figure 1: Flow chart for Bhaskara & AbuHamza –Shereen Method
- 5. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
144
Performance Measurements
Progressive Method Vs Bhaskara &Abu Hamza-Sheern Method
Table 1 :Performance measurement of 2 Methods.
Method 1 Method 2
(Progressive Method)
Input 60 Million
Barcodes Scanned
Images
Image processing
Through Assembly
Method.
Recommended Bhaskara
& Abu Hamza -Shereen
Method.
Decision Making
Capability
2 Minutes from time
Image(Barcode) is First
scanned. If first Robot
Breaks down Processing
times Goes down
1 minute from the time
Image (Barcode) is First
Scanned. If any robot
breaks down its work will
be distributed. Mobile
robots Process Images
parallel.
Average Time for Image
Processing
18 Minutes for Processing
30 Million Images With
Assumption That that all
mobile robots won’t break
or error out.
10 Minutes for Processing
30 million Images.
Message Loss or
Resource Allocation
Around 1 time Resource
lock or Object Lock for
every 1.5 Million
Images.If any of Robots
Breaks down Human
Intervention is needed
Results In Switch time,
Other Depended Mobile
idle time.
Resource lock or Object
Lock for every 2.7 Million
Images. Record locks are
resolved by Lead Robot by
scheduling and monitoring.
Number of Comparisons 35 Millions this Includes
Duplicate Images
Around 38 Million
Comparisons.
Assumptions and calculations
1) Input 30 millions Images (Barcodes) are converted into a DB2/400 file. This is done by
extracting the barcode image contents Through Translator this Table is created with columns
like barcode, shipper number, service type, destination terminal, delivery exceptions etc.
2) RPG language was used connect to db2/400 Database Located in Canada and is used to
simulate 2 Methods.
3) Method I is simulated by set of programs written ILERPG/400 which read Barcodes and
depending on business rules will populate output files. This process does not have Monitor
process. Whatever is processed by first program is read by second program and so on. If first
Program end up in error human intervention is needed and all other program process will be
halted till first program begins to run again.
- 6. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
145
4) Bhaskara &Abu-Hamza, Shereen Method are simulated by programs written in ILERPG
Language on AS/400 system. This program is supported by monitor program which
maintains and monitors starting and ending Relative record numbers of the work loads of all
Member mobile robots and in case if any of members fails or breaks down the Monitor
program will reassign that work load to other member how is done processing it’s assigned
work load.
The method which I am suggesting exhibits parallel processing. In case the Team lead robot
errors out the monitor program will resubmit another program which is capable of monitoring
other remaining Member robots.
5) If scanned images did not make it to Output file by predetermined time then there is
another program which calculates time taken for input image to get processed and make it to
output file and in case that time is missed then we generate an alert message. This will
measure time between Input obtained and processed download. Same formula is used to
measure 2 Methods.
Files Declaration
Figure 2: File Declaration
Data Structure Declaration
Figure 3: Data Structure Declaration
FIMGINPUT IF E K DISK
F
RENAME(SCANNED:IMPINP30)
FProcessfile IF A E K
DISK Prefix(scn)
Compile UF E K DISK
FRouting O E DISK
Download UF A E K DISK
D DS
D Barcode 1 14 0
D Location 1 4 0
D Destination 7 14 0
D Barcode Type 1 3 0
- 7. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
146
Figure 4: Relative Record Number Assignment Code. In this code each robot is assigned
equal work load.
Do while %notEof(IMGPF030)
if IMGSCAN = 'Y'
or IMGFOUND = 'Y'
or NOTINARRAY = 'Y'
or ROUNTINGS = 'Y'
RRN=0;
For I RRN to 60000 Step 1
Eval MR1RRN=RRN+1;
Endfor;
Eval RRN1=MR1RRN;
For I RRN1 to 120000 Step 1
Eval MR2RRN=MR2RRN+1;
Endfor;
Eval MR3RRN=MR2RRN;
For I MR3RRN to 180000 Step 1
Eval MR3RRN=MR3RRN+1;
Endfor;
For I MR3RRN to 240000 Step 1
Eval MR3RRN=MR3RRN+1;
Endfor;
EVAL MR4RRN=MR3RRN;
For I 1 to 60000 Step 1
Eval MR4RRN=MR4RRN+1;
Endfor;
on-error 01021
Endif
Enddo.
- 8. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
147
Figure 5: Image Processing jobs Running as a Team with Team Lead (TM1_MON) and
Team Members (MR1-MEM1, MR2-MEM2, MR3-MEM4
CONCLUSIONS
The objective of this paper is to prove that mobile robots when organized as a team
with designated team lead and designated team members will take care of any task assigned
in timely and highly efficient way with less resource consumed when compared to same
team of mobile robots which work as a team but without designated team lead and team
members.
Future Work
My future work deals with How to Make Mobile Robots communicate more efficiently.
Acknowledgements
I thank my guide Dr. Bhaskara Reddy, Associate Professor in the department of
computer Science and technology at Sri Krishnadevaraya University. I thank my colleague
Mohamed Farag, Post graduate student in Maharishi University of Management in Iowa. I
thank my parents, my wife and my son for their great support.
- 9. International Journal of Computer Engineering and Technology (IJCET), ISSN 0976-
6367(Print), ISSN 0976 – 6375(Online) Volume 4, Issue 3, May – June (2013), © IAEME
148
REFERENCES
1)http://robotics.iiit.ac.in/?n=Research.MobileRobotics
2)Military Robot http://www.faculty.rsu.edu/users/c/clayton/www/sasnett/paper.htm
3) http://www.iri.upc.edu/research/mobile_robotics
4)http://www.eetimes.com/design/industrial-control/4013686/Enhancing- automotive-
manufacturing-with-robots
5) http://www.industryweek.com/robotics/industrial-robotics-expands-beyond- automotive-
industry
6)http://www.robotics.org/content-detail.cfm/Industrial-Robotics-Feature-Article/Robots-in-
Aerospace-Applications/content_id/3591
7)http://www.wtec.org/robotics/report/05-Industrial.pdf
8)ftp://ftp.cc.gatech.edu/pub/people/arkin/archive/00059355.pdf
9)http://www.ndu.edu/es/programs/academic/industry/reports/2011/pdf/icaf-is-report-
robotics- autonomous-systems-2011.pdf
10)http://robotics.usc.edu/~maja/teaching/cs584/papers/JenningsKirkwoodWatts98.pdf
11)http://cres.usc.edu/pubdb_html/files_upload/50.pdf
12) Mr.J.Rajarajan and Dr.G.Kalivarathan, “Influence of Local Segmentation in the Context
of Digital Image Processing – A Feasibility Study”, International Journal of Computer
Engineering & Technology (IJCET), Volume 3, Issue 3, 2012, pp. 340 - 347, ISSN Print:
0976 – 6367, ISSN Online: 0976 – 6375.