Fuzzy and Neuro-fuzzy Based Co-operative Mobile robots

This chapter outlines the development of intelligent multi-agent robot teams that are capable of acting autonomously and of collaborating in a dynamic environment to achieve team objectives. A biologically-inspired collective behavior for a team of co-operating robots is proposed. A modification of the subsumption architecture is also proposed for implementing the control of the individual robots. The chapter also presents a fuzzy logic technique to enable the resolution of conflicts between contradictory behaviors within each robot. It then discusses a neuro-fuzzy based adaptive action selection architecture that enables team of robot agents to achieve adaptive cooperative control to perform two proof-of-concept co-operative tasks: dynamic target tracking and boxpushing. Simulated and real experiments have been conducted to validate the proposed techniques. These experiments are illustrated in the chapter and they prove that the use of fuzzy logic enables the resolution of conflicts between contradictory behaviors by selecting an action that represents the consensus among the behaviors and that best satisfies the decision objectives encoded in them. © 2006