Variable composition based adaptive fuzzy control of double inverted pendulum

Rule number explosion and adaptive weights tuning are two main issues in the design of fuzzy control systems. To overcome the problems, we propose the method of state variable composition, and combine it with adaptive control scheme to tune controller parameters. Furthermore, LQR feedback controller coefficients of double inverted pendulum are used as factors of state variable composition to design the proposed ANFIS controller. Numerical simulation and real-time experiment results on the tracking performance, disturbance resisting capability and robustness against model parameter perturbance show that the proposed method yields better control quality.