Genetic algorithm based fully automated and adaptive fuzzy logic controller

In this paper, an integration of fuzzy logic controller (FLC) and genetic algorithm (GA) is developed with a view to make the design process fully automatic, without requiring any human expert knowledge. Here, GA is used in two stages simultaneously: the first stage involves selection and definition of fuzzy rules, while the second stage performs an optimal selection of membership function types associated to the fuzzy rules. It is argued that the performance of an FLC greatly depends on the fuzzy rules as well as the types of membership functions associated to the fuzzy sets. Thus, the aforementioned two-stage GA is a viable solution for designing an efficient FLC system. In order to evaluate performance, the proposed approach is applied to a well-known benchmarking controller design task, "backing up a truck reversing system". The simulation result exhibits superior performance and thereby validates the proposed integrated GA and FLC system.