Path Tracking Control of Intelligent Vehicles Based on Disturbance Rejection Model Predictive Control

In response to the issue of reduced trajectory tracking accuracy in intelligent vehicles when navigating complex curved roads due to environmental and vehicle model simplification disturbances, an active steering control method based on Active Disturbance Rejection Control Model Predictive Control (ADRC-MPC) is proposed. Initially, disturbance rejection control is designed, and state estimators are employed to estimate uncertainty parameters, thereby mitigating the impact of model simplifications and external uncertainties. Subsequently, Model Predictive Control (MPC) is combined to optimize trajectory tracking accuracy within the prediction horizon. Finally, feedforward compensation is applied to the control input using the estimated disturbance values, resulting in the optimal control input for the vehicle under dynamic disturbances. Simulation results indicate that, under complex curved road conditions, the proposed method maintains good control performance, with improved path tracking accuracy, driving stability, and robustness.