Generating fuzzy rules by genetic algorithms

A general method is developed to generate fuzzy rules by using genetic algorithms (GAs) and a fuzzy logic controller (FLC). By using GAs as a learning procedure and a FLC as the system's performance evaluator, the proposed architecture can construct an input-output mapping in the form of fuzzy if-then rules. The performance of the new architecture is compared with an artificial neural networks controller and pure limited-rule fuzzy rule controller for the truck back-upper problem.<
>