Design of fuzzy-learning fuzzy controllers

In this paper, the nested design approach with partial fuzzy if-then rule base is proposed to reduce the complexity of the fuzzy controllers. In the design procedure, the fuzzy controllers are first constructed with basic fuzzy if-then rules. When the performance requirements of the fuzzy control system are not satisfied the fuzzy controller will be adjusted accordingly. Based on the nested structure of the partial fuzzy rule base, the fuzzy rule base of a fuzzy controller is modified by replacing the unsatisfactory fuzzy rules with additional efficient fuzzy rules. Each fuzzy region of the input space can be partitioned all by itself into fuzzy subregions to avoid generating the redundant rules. And the nested structure of the partial fuzzy rule base can be further simplified into the nonnested one. Finally, simulations are carried out to show the effectiveness of the fuzzy controllers with nested fuzzy if-then rules.