Particle Swarm Optimization Based Adaptive Strategy for Tuning of Fuzzy Logic Controller

This paper presents a new method for learning and tuning a fuzzy logic controller automatically by meansof a particle swarm optimization (PSO). The proposed self-learning fuzzy logic control that uses the PSOwith adaptive abilities can learn the fuzzy conclusion tables, their corresponding membership functions andfitness value where the optimization only considers certain points of the membership functions. To exhibitthe effectiveness of proposed algorithm, it is used to optimize the Gaussian membership functions of thefuzzy model of a nonlinear problem. Moreover, in order to design an effective adaptive fuzzy logiccontroller, an on line adaptive PSO based mechanism is presented to determine the parameters of the fuzzymechanisms. Simulation results on two nonlinear problems are derived to demonstrate the powerful PSOlearning algorithm and the proposed method is able to find good controllers better than neural controllerand conventional controller for the target problem, cart pole type inverted pendulum system.