Design of ANFIS controller based on fusion function for rotary inverted pendulum

Rule number explosion and adaptive weights tuning are two main issues in the design of fuzzy control systems. To overcome the problems, a method has been used for control of the rotary inverted pendulum (RIP) using linear fusion function based on LQR mapping, and combine it with adaptive control scheme to tune controller parameters using ANFIS. By using fusion, the output variables of the system with four dimensions are synthesized as two variables: error and variation of error. The method is applied to the approximate linear model, and the experimental results show that the applied method has better tracking performance, disturbance resisting performance, and robustness against model parameter perturbation.