On Extended State Sequence Prediction Based MPC Path Tracking Control for Autonomous Vehicle

Path tracking control is a critical technology for autonomous vehicles, yet it faces significant challenges due to dynamic disturbances caused by varying road conditions and model uncertainties. To address these issues, this paper proposes an extended state sequencebased model predictive controller (PESO-MPC) for vehicle path tracking. First, an extended state observer (ESO) is developed to estimate dynamic disturbances in real time, and is coupled with the online identification of the disturbance prediction model to obtain disturbance sequences. Subsequently, an enhanced MPC framework incorporating the extended state sequence into the prediction model is established. By solving a quadratic programming problem, a control law with dynamic disturbance rejection capability is derived. The simulation and real vehicle experiment results demonstrate that PESO-MPC has superior performance, reducing the root mean square error(RMSE) by over <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{4 1. 2 4 \%}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 8. 6 4 \%}$</tex> compared to conventional Model Predictive Control (MPC) and Nonlinear MPC (NMPC), respectively.