Title: A Software Architecture and Tools for Autonomous Robots that Learn on Mission
1A Software Architecture and Tools for Autonomous
Robots that Learn on Mission
K. Kawamura, M. Wilkes, A. Peters, D.
Gaines Vanderbilt UniversityCenter for
Intelligent Systems Jet Propulsion
Laboratory http//shogun.vuse.vanderbilt.edu/cis/D
ARPA/
February 2002 MARS PI Meeting
2Objective
- Develop a multi-agent based robot control
architecture for humanoid and mobile robots that
can - accept high-level commands from a human
- learn from experience to modify existing
behaviors, and - share knowledge with other robots
3Accomplishments
- Multi-agent based robot control architectures for
humanoid and mobile robots have been developed - Agent-based Human-Robot Interfaces have been
developed for humanoid and mobile robots - SES (Sensory EgoSphere), a short-term robot
memory, was developed and transferred to NASA/JSC
Robonaut group - SES- LES (Landmark EgoSphere)- based navigation
algorithm was developed and tested - SES knowledge sharing among mobile robots was
developed and tested - SAN-RL (Spreading Activation Network -
Reinforcement Learning) method was applied to
mobile robots for dynamic path planning
4Presentation / Demo
- Multi-agent based Robot Control Architecture
- Humanoid
- Mobile robots
- Agent-based Human Robot Interfaces
- Humanoid (face-to-face)
- Mobile robots (GUI)
- Sensory EgoSphere (SES)
- Humanoid
- Mobile robots
- SES and LES based Navigation
- SES and LES Knowledge Sharing
- Dynamic Path Planing through SAN-RL
5Multi-Agent Based Robot Control Architecture for
Humanoids
Novel Approach Distributed architecture that
expressly represents human and humanoid internally
Publication 1,2
6Multi-Agent Based Robot Control Architecture for
Mobile Robots
Novel Approach Distributed, agent-based
architecture to gather mission relevant
information from robots
Publication 7
7Agent-based Human-Robot Interfaces for Humanoids
- Self Agent (SA)
- monitors humanoids activity and performance for
self-awareness and reporting to human - determines the humanoids intention and response
and reports to human
- Human Agent (HA)
- observes and monitors the communications and
actions of people - extracts persons intention for interaction
- communicates with people
Novel Approach Modeling the humans and
humanoids intent for interaction
Publication 3,4,5
8Agent-based Human-Robot Interface for Mobile
Robots
Camera UIC
Sonar UIC
Novel Approach Interface that adapts to the
current context of the mission in addition to
user preferences by using User Interface
Components (UIC) and an agent-based architecture
Publication 7
9Sensory EgoSphere (SES) for Humanoids
- Objects in ISACs immediate environment are
detected - Objects are registered onto the SES at the
interface nodes
closest to the objects perceived locations - Information about a sensory object is stored in a
database with the node location and other index
Publication 2
10Sensory EgoSphere Display for Humanoids
Provides a tool for person to visualize what
ISAC has detected
11Sensory EgoSphere (SES) for Mobile Robots
- The SES can be used to enhance a graphical user
interface and to increase situational awareness - In a GUI, the SES translates mobile robot
sensory data from the sensing level to the
perception level in a compact form - The SES is also used for perception-based
navigation with a Landmark EgoSphere - The can be also used for supervisory control of
mobile robots - Perceptual and sensory information is mapped on a
geodesically tessellated sphere - Distance information is not explicitly
represented on SES - A sequence of SESs will be stored in the database
SES
2d EgoCentric view
Top view
Publication 6
12SES- and LES-Based Navigation
- Navigation based on EgoCentric representations
- SES represents the current perception of the
robot - LES represents the expected state of the world
- Comparison of these provide the best estimate
direction towards a desired region - more
Future Work
Novel Approach Range-free perception-based
navigation
Publication 8
13SES and LES Knowledge Sharing
- Skeeter creates SES
- Skeeter finds the object
- Skeeter shares SES data with Scooter
- Scooter calculates heading to the object
- Scooter finds the object
- Scooter has the map of the environment
- Scooter generates via LESs
- Scooter shares LES data with Skeeter
- Skeeter navigates to the target using PBN
Novel Approach A team of robots that share SES
and LES knowledge
Future Work
Publication 9
14Dynamic Path Planning through SAN-RL(Spreading
Activation Network - Reinforcement Learning)
High level command with multiple goals
- Behavior Priority
- Using the shortest time
- Avoid enemy
- Equal priority
- More
Get initial data from learning mode
After finish training send data back to DB
SAN-RL
activate/deactivate robots behaviors
Scooter
Atomic Agents
Novel Approach Action selection with learning
for the mobile robot
Publication 10
15Publications
- K. Kawamura, R.A. Peters II, D.M. Wilkes, W.A.
Alford, and T.E. Rogers, "ISAC Foundations in
Human-Humanoid Interaction", IEEE Intelligent
Systems, July/August 2000. - K. Kawamura, A. Alford, K. Hambuchen, and M.
Wilkes, "Towards a Unified Framework for
Human-Humanoid Interaction", Proceedings of the
First IEEE-RAS International Conference on
Humanoid Robots, September 2000. - K. Kawamura, T.E. Rogers and X. Ao, Development
of a Human Agent for a Multi-Agent Based
Human-Robot Interaction, Submitted to First
International Joint Conference on Autonomous
Agents and Multi-Agent Systems (AAMAS 2002),
Bologna, Italy, July 15-19, 2002. - T. Rogers, and M. Wilkes, "The Human Agent a
work in progress toward human-humanoid
interaction" Proceedings 2000 IEEE International
Conference on Systems, Man and Cybernetics,
Nashville, October, 2000. - A. Alford, M. Wilkes, and K. Kawamura, "System
Status Evaluation Monitoring the state of agents
in a humanoid system, Proceedings 2000 IEEE
International Conference on Systems, Man and
Cybernetics, Nashville, October, 2000. - K. Kawamura, R. A. Peters II, C. Johnson, P.
Nilas, S. Thongchai, Supervisory Control of
Mobile Robots Using Sensory EgoSphere, IEEE
International Symposium on Computational
Intelligence in Robotics and Automation, Banff,
Canada, July 2001. - K. Kawamura, D.M. Wilkes, S. Suksakulchai, A.
Bijayendrayodhin, and K. Kusumalnukool,
Agent-Based Control and Communication of a Robot
Convoy, Proceedings of the 5th International
Conference on Mechatronics Technology, Singapore,
June 2001. - K. Kawamura, R.A. Peters II, D.M. Wilkes, A.B.
Koku and A. Sekman, Towards Perception-Based
Navigation using EgoSphere, Proceedings of the
International Society of Optical Engineering
Conference (SPIE), October 28-20, 2001. - K. Kawamura, D.M. Wilkes, A.B. Koku, T.
Keskinpala, Perception-Based Navigation for
Mobile Robots, accepted Proceedings of
Multi-Robot System Workshop, Washington, DC,
March 18-20, 2002. - D.M. Gaines, M. Wilkes, K. Kusumalnukool, S.
Thongchai, K. Kawamura and J. White, SAN-RL
Combining Spreading Activation Networks with
Reinforcement Learning to Learn Configurable
Behaviors, Proceedings of the International
Society of Optical Engineering Conference (SPIE),
October 28-20, 2001.
16Acknowledgements
- This work has been partially sponsored under the
- DARPA MARS Grant DASG60-99-1-0005
- and from the
- NASA/JSC - UH/RICIS Subcontract NCC9-309-HQ
- Additionally, we would like to thank the
following CIS students - Mobile Robot Group Bugra Koku, Carlotta Johnson,
Turker Keskinpala, Anak Bijayendrayodhin, Kanok
Kusumalnukool, Jian Peng - Humanoid Robotic GroupTamara Rogers, Kim
Hambuchen, Christina Campbell