Circular Arc Extension to Pure Pursuit for Accurate End-of-Path Tracking

The pure pursuit algorithm, featuring simple computation and strong robustness, has been widely applied in lateral control of autonomous driving. However, when the distance between the vehicle and the destination is less than the look-ahead distance, the classical pure pursuit algorithm encounters the problem of being unable to find a look-ahead point on the trajectory. This causes significant heading errors when the autonomous vehicle stops at the end of a curved trajectory. To address this issue, this paper proposes a method for extending the virtual trajectory with a circular arc. This method generates a virtual arc segment at the end of the original trajectory, considering both the vehicle’s position and orientation as well as that of the original trajectory endpoint, thereby enabling a more accurate look-ahead point at a specified look-ahead distance. Extensive simulation experiments demonstrate that the proposed method significantly improves tracking accuracy at end of the trajectory while maintaining the simplicity of the pure pursuit, outperforms existing approaches.