Development of a hybrid crisp-fuzzy logic algorithm optimised by genetic algorithms for path-planning of an autonomous mobile robot

Abstract

This paper deals with the problem of autonomous mobile robot navigating in unknown environments. The approach developed is reactive in a behaviour-based framework aimed at multiple obstacle avoidance. We use a novel hybrid logic integrating fuzzy and crisp to enable the robot with a human like decision-making ability and avoid failure when world set-ups mismatch with the fuzzy system designer expectancies. Finally, a genetic algorithm is used as means of optimised search within the space of the fuzzy parameters of the hybrid system.