Guard-Robo : Multi-Robot Learning in a Cooperative Guarding Task

Abstract of project:

Cooperative Robotics has drawn a great research interests among the robotics professionals in recent days. This defines a system where multiple autonomous mobile robots exhibit cooperative behaviors. Groups of mobile robots are constructed, with an aim to studying such issues as group architecture, re source conflict, origin of cooperation, learning, and geometric problems. As yet, few applications of cooperative robotics have been reported, and sup porting theory is still in its formative stages. In this project we describe such a situation which is inherently multi-agent and cooperative-guarding an area by a group of mobile robots. The mobile robots uses a negotiation based approach to position themselves at the boundary of guarding area, covering the maximum perimeter of the guarding place. They also exhibits some behaviors to recognize the intruders to that guarding area, and then try to stop them to enter inside the guarding area. All these behaviors are stimulated based on the dynamically changing environment, and best behavior is choosen based on the negotiation between all the robots. This negotiation based protocol is a common generic protocol that comes from negotiation based game theoretic approach, and runs independently at each individual agents, but finally comes to an convergence state, where all the agents have achieved their goals.

This entry was posted in 2010 Course Projects and tagged , , , , . Bookmark the permalink.

Comments are closed.