A trajectory tracking control method for autonomous vehicle based on adaptive model predictive control

Abstract

In order to further enhance the accuracy, timeliness, and adaptability of trajectory tracking for autonomous vehicles, an innovative Adaptive Model Predictive Control (AMPC) algorithm is proposed. Building upon the traditional Model Predictive Control (MPC) control algorithm, this approach takes the current vehicle speed and reference trajectory curvature as system inputs and utilizes a fuzzy control algorithm to achieve real-time adjustment of the prediction horizon for the MPC trajectory tracking controller. Through multiple simulation experiments and employing a third-degree polynomial fit, the control law for the control horizon is obtained, enabling adaptive adjustments. The reliability of the proposed Adaptive MPC trajectory tracking control method is validated through joint simulation in Matlab/Simulink and CarSim. The research results demonstrate that, compared to traditional MPC control, Adaptive MPC control significantly improves the accuracy of vehicle trajectory tracking, reduces average computation time, and provides a novel and effective approach for trajectory tracking control in autonomous vehicles.