Adaptive trajectory tracking control for autonomous vehicles based on heading angle deviation

Jiaqing Hao, Yuhan Guo, Yuqiong Wang, Xinyi Xu, Huimin Cai

Precise trajectory tracking holds significant importance for autonomous vehicle navigation. Conventional lateral control strategies are oriented towards eliminating yaw angle deviation. However, this study reveals that such strategies are inadequate under high-speed and high-curvature conditions, where non-zero steady-state yaw angle deviation exists. An enhanced lateral tracking strategy is put forward.Initially, a trajectory tracking error model is constructed, and a steady—state error analysis is conducted. The results demonstrate that heading angle deviation is a more superior lateral tracking index than yaw angle deviation. Subsequently, a feedforward-feedback controller utilizing heading angle deviation is designed. This controller is compared with a yaw-angle-based controller and an Active Disturbance Rejection Controller (ADRC) strategy based on global lateral displacement.Co-simulations in CarSim/Simulink for circular turning and single lane-change yield two notable contributions: 1) During lane-change maneuvers, the heading-angle-based controller reduces the peak lateral displacement error by approximately 83.3% in comparison to the yaw-angle-based controller, thereby enhancing tracking accuracy. 2) A performance mapping between controller types and path geometries is established. The ADRC strategy exhibits better performance on non-return paths, while the heading-angle-based controller guarantees reliable tracking on return paths.This research offers a theoretical foundation and a practical framework for the selection of adaptive, high-precision trajectory tracking strategies in specific driving scenarios.