Acquisition of fuzzy control rules for a mobile robot using genetic algorithm

Abstract

Fuzzy controls have been widely used in industry for its high degree of performance in human-computer interactions. DNA coding method, which is one of the coding methods in genetic algorithm, is based on biological DNA and a mechanism of development from the artificial DNA. This method has redundancy and overlapping of genes, and it is suitable for knowledge representation. In this paper, we propose the parallel genetic algorithm using the DNA coding method. This paper applies this method to acquisition of fuzzy control rules with multiple input/output system for a mobile robot. This method can select input variables from many candidates and tune membership functions. The result of simulation shows that the robot can reach the goal quickly and efficiently. Effective fuzzy rules for the mobile robot are acquired by using this method while the length of the chromosomes in the population is automatically adjusted.