Hierarchical fuzzy controller for a nonholonomic mobile robot

Abstract

The objective of this paper is to find optimal trajectories of a mobile robot during its motion from an initial position to a desired one in an autonomous way. For that, a new approach which uses hierarchical fuzzy systems has been proposed. This approach combines fuzzy logic systems and the potential field method. Firstly, the developed controller is based on fuzzy logic systems: this controller sends the speed values of each driving wheel to the robot, while seeking the goal. Secondly, the obstacle avoidance behavior is inspired from the potential field method: the robot is attracted to the target and pushed from obstacles. Genetic algorithms are used to optimize the consequences of fuzzy rules. Simulation results illustrate the validity and the efficiency of the suggested approach.