A new approach in self-generation of fuzzy logic controller by means of genetic algorithm

Ever since its first development, Fuzzy Logic Controllers (FLC) have been popular among the practitioners due to its robustness, interpretability, and especially its ability to handle imprecision. Many constructions of these controllers are still heavily dependent on the presence of experts' knowledge. This drawback has been investigated by many researchers, resulting in several methods integrated into the construction of FLC. This paper presents a novel method to generate FLC with the minimum involvement of experts. The method integrates Genetic Algorithm (GA) into the design process. The concept of gene pool, rule filter, and two levels of encoding were devised for the method. Simulative tests to control a nonlinear single tank system and a benchmark inverted pendulum system were undertaken. The results show the performance of the proposed method to create FLCs for both systems with minimum need of prior knowledge.