Research on Adaptive Optimization Trajectory Tracking Algorithm based on MPC

Abstract

Aiming at the problems of large trajectory tracking error and poor adaptability to different road conditions of the traditional constant predictive time-domain model predictive control algorithm, a new variable predictive time-domain adaptive trajectory tracking controller is proposed by fusing the traditional model predictive control algorithm and fuzzy control algorithm. The fuzzy control algorithm is used to adjust the predictive time-domain required by the predictive model controller according to the two parameters of vehicle speed and ground adhesion coefficient. The Gaussian membership function is used to formulate fuzzy rules. Through the operation of fuzzy control rules, the predictive time-domain parameters matching the current system are automatically output and transmitted to the model predictive controller to control the vehicle for trajectory tracking to improve the adaptability of trajectory tracking to different road conditions. Simulink and CarSim joint simulation show that the comprehensive use of the proposed adaptive trajectory tracker can effectively improve the accuracy and stability of vehicle trajectory tracking on low adhesion wet and slippery roads.