Evolutionary-based learning applied to fuzzy controllers

Fuzzy logic controllers constitute knowledge-based systems that include fuzzy rules and fuzzy membership functions to incorporate the human knowledge into their knowledge base. The definition of fuzzy rules and fuzzy membership functions is actually affected by subjective decisions, having a great influence over the whole FLC performance. Some efforts have been made to obtain an improvement on system performance by incorporating learning mechanisms to modify rules and/or membership functions. Genetic algorithms (GAs) are probabilistic search and optimization procedures based on natural genetics. This paper proposes a new way to apply GAs to FLCs, and applies it to a FLC designed to control the synthesis of biped walk of a simulated 2-D biped robot.<
>