1. SECURE: Semantics Empowered
resCUeenviRonmEnt
demo @ SSN-ISWC2011
P. Desai, C. Henson, P. Anandtharam, A. Sheth
Ohio Center of Excellence in Knowledge-Enabled
Computing (Kno.e.sis),
Wright State University, Dayton, Ohio
Semantic Sensor Web @ Kno.e.sis
3. Motivation
– Environment ignorant
• Machines without any sensors
http://www.familycourtchronicles.com/philosophy/dissonance/remote-control-car.jpg
4. Motivation
– Environment sensing
• Machines with sensors
Photo courtesy NASA
The autonomous Urbie is designed for various urban operations,
including military reconnaissanceand rescue operations.
5. Motivation
– Environment comprehending
• Machine with sensors + perceiving background
knowledge + comprehending background knowledge
Traffic signals
Google’s car that won
the DARPA challenge. pedestrians
Stimuli and others
on roads.
Speed restrictions
http://www.nytimes.com/2010/10/10/science/10google.html
6. Project Focus
• Building a rescue robot (mobile-platform) with
many sensors.
• Data Collection and annotation using SSN
ontology.
• Analysis to be carried out for situational
awareness using perception ontology [2].
• Visualization of the emergency situation in
terms of abstractions.
7. System Architecture
Robot (Mobile Platform) With Sensors
Data Collection
Events in
environment
Annotation Visualization
Paper on Fire
Perceptual
Analysis
8. Data Collection + Annotation
• Collection of data from sensors on the robot.
– Position data: Observation form position sensors.
– Sensor data: Observation from environment
sensors.
• Annotation of raw sensor data.
– Use of SSN ontology which has concepts to
describe sensors and their observations.
Raw Sensor Data
Robot (Mobile Platform) With Sensors Position data
Position
Data Stream
Sensor Data (CO2,
Temperature, IR, CO
data.)
Annotated Annotated
Paper on Fire Data (triple Data Stream
store)
9. Perceptual Analysis
• Perceptual ontology used to derive
abstractions from annotated sensor data.
• Domain knowledge is used to derive these
abstractions.
Annotated
Sensor Data
Abstraction
Stream
Perceptual
Reasoning
Domain
Knowledge
Images: http://static3.depositphotos.com/1001416/130/i/950/depositphotos_1304999-Sheet-of-the-old-scorched-paper-and-fire.jpg
http://www.firesystems.net/images/portable-fire-extinguishers/types-1.jpg
http://www.blogcdn.com/www.engadget.com/media/2007/02/irobot-packbot-510.jp
10. Visualization
• Visualization serves as a dashboard for
presenting real-time:
– Raw sensor data
– Position Data
– Derived abstractions
– Video of the robot
11. Demo
SECURE
Online
Demo:
http://www.youtube.com/
watch?v=smu9mPFFyNs
Local
12. Conclusions
• Robot with perceptual abilities give out
abstractions that are intuitive to humans.
• Demonstrated a real-time physical system that
uses domain knowledge to process
heterogeneous sensor data.
• Demonstrated visualization of events (as
abstractions) in an emergency situation in
real-time.
13. References
[1] Cory Henson, KrishnaprasadThirunarayan, AmitSheth, Pascal Hitzler,
'Representation of Parsimonious Covering Theory in OWL-DL,' In: Proceedings of
the 8th International Workshop on OWL: Experiences and Directions (OWLED
2011), San Francisco, CA, United States, June 5-6, 2011.
[2] Cory Henson, KrishnaprasadThirunarayan, AmitSheth. An Ontological Approach to
Focusing Attention and Enhancing Machine Perception on the Web. Applied
Ontology, 2012. (accepted).
Demos, Papers and more at: http://semantic-sensor-web.com
Semantic Sensor Web @ Kno.e.sis
Hinweis der Redaktion
Machines interpreting the sensor data and sending only abstractions to the humans would be ideal. Else, the human has to interpret the sensor data which is not intuitive for humans.
Note: No free food! Someone has to do the processing!Machines/HumansIf traffic signal is red, then stop.If there are pedestrians crossing the road, then stop.If the speed limit is less than the current speed, slowdown.