Improved Pure Pursuit Algorithm Based Path Tracking Method for Autonomous Vehicle
Meng Wang, Xue Lv, Juexuan Chen, Xiaocong Su
Abstract
Pure pursuit algorithm is commonly used in path tracking control of autonomous vehicle for its high real-time performance. Due to the problem of “taking shortcuts,” traditional pure pursuit algorithms usually have the problem of low path tracking accuracy in curved road scenarios. To address the issue, a path tracking control method based on improved pure pursuit algorithm is proposed. This method builds upon traditional pure pursuit theory and dynamically adjusts the look-ahead distance based on vehicle speed and road curvature radius information, allowing it to adapt to different road scenarios. This effectively addresses the problem of large path tracking errors in curved road scenarios. Furthermore, a fuzzy feedback control is employed to compensate for control variable and enhance tracking accuracy across various scenarios. Simulations and real-world experiments demonstrate that the proposed method significantly improves path tracking accuracy compared to traditional pure pursuit methods, particularly in curved road scenarios. The maximum lateral deviation is reduced by over 50%, realizing the precise tracking of autonomous vehicle on the park roads.