关键词: 高斯过程回归, 模型预测控制, 轨迹跟踪, 无人驾驶

Abstract: Trajectory tracking is a crucial functionality in the autonomous driving control system. Acknowledging the critical role of the vehicle dynamic model, this paper proposes a method utilizing model predictive control with state compensation from Gaussian process regression for path tracking. First, a vehicle state fusion estimation method based on the single-track dynamics model was developed to obtain the GPR compensation model. Second, based on the trajectory tracking error from vehicle dynamics, the iterative equation for GPR error compensation within the predictive horizon is derived to dynamically compensate for model errors in the vehicle state prediction. Finally, a real-vehicle validation platform is constructed to validate the method under typical driving conditions. Compared with predictive control methods without GPR compensation, the results show the proposal method significantly improved trajectory tracking accuracy while simultaneously enhancing vehicle comfort.

Key words: Gaussian process regression, model predictive control, trajectory tracking, autonomous driving