Online ANFIS controller based on RBF identification and PSO

Abstract

Adaptive neuro-fuzzy inference system (ANFIS) is combining a neural network with a fuzzy system results in a hybrid neuro-fuzzy system, capable of reasoning and learning in an uncertain and imprecise environment. In this paper online training of ANFIS is done using radial basis function (RBF) neural network. In this online approach, identification of controlled plant is done, and based on this identification, the weights and coefficients are adjusted timely. Finally, to overcome initialization problem, using Particle swarm optimization (PSO) as an evolutionary algorithm is proposed.