Analyzing the Impact of Wind on Trajectory Tracking for Unmanned Vehicles Based on Road Adhesion Coefficient Estimation

Abstract

This paper presents a novel approach to address the challenge of trajectory tracking control in unmanned vehicles (UVs) operating on roads with varying ad hesion coefficients and subject to wind disturbances. The proposed method integrates real-time estimation of road adhesion coefficients with a wind disturbance model, embedding both into the trajectory tracking control system. Specifically, the Unscented Kalman Filter (UKF) is employed to dynamically estimate the road adhesion coefficients. These estimates are then combined with the wind disturbance model and incorporated into a Model Predictive Control (MPC) framework, enhancing the robustness and accuracy of trajectory tracking under varying conditions. The integrated simulation results from CarSim/Simulink demonstrate high accuracy in road adhesion coefficient estimation for trajectory tracking control, with tracking errors consistently remaining below 0.1 across different road adhesion conditions. For example, the road adhesion coefficient estimation error for the docking surface is 0.5%, and for the split road surface, it is 0.75%, indicating the approach’s applicability across a wide range of road conditions. Notably, when the wind speed reaches level-five intensity and a shelter-type vehicle is used as a test case, the proposed controller significantly outperforms the controller that does not consider wind effects. Under both high and low adhesion conditions, the controller reduces the yaw angle by approximately 15% and 30%, respectively, and the yaw rate by approximately 25% and 20%. These results provide strong evidence for the effectiveness of the proposed trajectory tracking controller in enhancing the stability and control performance of unmanned vehicles.