Combined lateral and longitudinal trajectory tracking control of intelligent vehicle based on adaptive RMPC algorithm

In order to improve accuracy and adaptability of intelligent vehicle trajectory tracking under the condition of severe environment and time-variant parameters, the combined lateral and longitudinal trajectory tracking control strategy based on adaptive robust model predictive control (RMPC) algorithm is proposed. Firstly, the vehicle dynamics model with lateral and longitudinal coupling characteristics is established. Secondly, the robustness optimization function is constructed in min-max form according to the lateral and longitudinal tracking error as well as yaw angle error corresponding to the reference trajectory. Linear matrix inequality (LMI) theory is used to solve the optimization problem. Then, the dynamic adjustment strategy of the weight coefficient matrix by the fuzzy controller is proposed according to the trajectory tracking error, so as to improve the adaptability of controller and the accuracy of trajectory tracking. Finally, the effectiveness of controller is verified by Simulink/CarSim co-simulation and hardware-in-loop (HiL) test under different speeds and parameters perturbation. The results show that the combined lateral and longitudinal trajectory tracking control strategy proposed in this paper reduces the lateral error by 0.4 m and the heading error by 7° compared with the decentralized control, which can achieve more accurate and robust tracking of the target trajectory.