Active MPC for High-Precision Lateral Trajectory Tracking of Autonomous Vehicles Against Longitudinal Velocity Disturbances

Aiming at the problem that the lateral trajectory tracking accuracy of autonomous vehicles degrades due to the influence of longitudinal velocity disturbances during driving, this paper designs a lateral trajectory tracking controller based on Model Predictive Control (MPC). The controller is established on the basis of a three-degree-of-freedom (3-DOF) vehicle dynamics model that incorporates lateral position, heading angle, lateral velocity and yaw rate, and a constant crosswind disturbance term is integrated into the model to fit the actual driving conditions. In addition, the continuous-domain dynamics model is discretized to lay a foundation for the online solution of MPC. To verify the performance of the designed controller, a continuous reference trajectory generated by a combination of sine and cosine functions is constructed, and a longitudinal velocity disturbance that follows a normal distribution and is processed by a moving average filter is introduced in the simulation. The simulation results show that the proposed MPC controller can effectively suppress the negative impact of longitudinal velocity disturbances on lateral tracking performance and achieve high-precision tracking on the sinecosine composite reference trajectory, with the maximum lateral tracking error not exceeding 0.1 m and the steady-state error approaching zero. At the same time, the front wheel steering angle control input output by the controller changes smoothly without drastic fluctuations and strictly meets the preset constraint conditions. This research provides a reliable control scheme for the lateral trajectory tracking of autonomous vehicles in dynamic disturbance environments and improves the robustness and accuracy of vehicle trajectory tracking.