An Algorithm for Tightly Tracking the Trajectory

In some environments, such as in a narrow road, robots need track the target path tightly and run smoothly, keeping deviations within the required range. However, pure-pursuit algorithm as a common path tracking algorithm uses a constant guidance distance, which cannot satisfy these requirements. In this paper, we propose a tracking algorithm to solve this problem. Comparing with the pure-pursuit, the algorithm dynamically selects the guidance distance based on environments and makes an angle error compensation when calculating the steering radius. According to the result of the experiment, the tracking algorithm can better follow the target path and adapt to the environment.