Human-analogous Network-based Fuzzy logic control : A case study of servo control of a cart moving on a linear track

In this paper we present a new method to tune the rule base in a fuzzy logic controller using a neural network approach trained by human data, with application of servo-control of a cart moving on a linear track. We take advantage of direct implementation of human data into this Adaptive-Network-based Fuzzy Inference System (ANFIS) to optimize the number and type of the membership functions in the rule base and to tune the parameters associated with the antecedent and consequent of each fuzzy rule. We show through simulation and experiments that this controller outperforms a conventional controller based on Linear Quadratic Regulators (LQR).