Data‐driven generalized predictive control for car‐like mobile robots using interval type‐2 T‐S fuzzy neural network
Yi Shui, Tao Zhao, Songyi Dian, Yi Hu, Rui Guo, Shengchuan Li
Abstract
Abstract In this paper, for the nonlinear influencing factors such as uncertainty and external disturbance existing in the actual driving process of the car‐like mobile robot (CLMR), a data‐driven generalized predictive control (GPC) method based on interval type‐2 T‐S fuzzy neural network (IT2TSFNN) is proposed for the trajectory tracking of CLMR. The controlled auto‐regressive integrated moving average (CARIMA) model of the mobile robot is established by analyzing data samples and using IT2TSFNN. Then, a generalized predictive controller is designed for the CARIMA model. Also, the global convergence of IT2TSFNN is verified by the Stone‐Weirstrass theorem. Unlike most previous results, the proposed method does not rely on the mathematical model of the mobile robot but only on the historical data of its operation. Finally, the simulation results show that the proposed method can avoid repeatedly debugging parameters, deal with the influence of uncertain factors, and obtain higher tracking accuracy.