A Behavior-Based Controller with Evolutionary Adapted Fuzzy Rule Base for a Car-like Mobile Robot

摘要

This paper presents the conception of a behavior-based controller for a car-like mobile robot. The proposed controller consists of a set of behaviors and a coordinator. Each behavior is an independent fuzzy-logic component, which has its own input/output variables and fuzzy rule base. In order to select the active behavior in an appropriate situation, a specific coordinator is implemented. An evolutionary adaptation of the fuzzy rule base is proposed to adapt the robot to its environment to have more autonomy and robustness.