Automatic Vehicle Trajectory Tracking Control Based on Adaptive Model Prediction

Chen Wen-Jue,Liu Fu-Chao,Liu Ning,Zhao Xu,Zhao Hui
DOI: https://doi.org/10.23919/ccc63176.2024.10662069
2024-01-01
Abstract:In order to improve the track tracking accuracy and stability of unmanned vehicles with independent four-wheel steering and independent drive during emergency steering at high speed and low adhesion conditions, a track tracking and stability control method of unmanned vehicles based on adaptive model prediction was proposed. The method can adjust the lateral position error and yaw velocity error weight coefficient in real time according to the vehicle driving state. At the same time, in order to deal with the uncertainty interference in actual driving, the closed-loop feedback item is designed and the model predictive control is combined to improve the robustness of the whole vehicle system. Through Carsim and MATLAB/Simulink co-simulation, the vehicle’s track tracking and yaw stability performance were tested under high-speed, low adhesion and dual alignment conditions. The simulation results show that compared with the classical model predictive control method, the average lateral error of the proposed method is reduced by $34.4 \%$, the peak phase of yaw velocity is reduced by $54.7 \%$, and the peak of centroid side deflection Angle is reduced by 58.11%. It is proved that this method can improve vehicle tracking accuracy and ensure vehicle stability under high speed and low adhesion conditions, and optimize vehicle driving stability and safety.
What problem does this paper attempt to address?