A Hierarchical Vehicle Motion Planning Method For Cruise In Parking Area

Unstructured roads, as a scene that autonomous vehicles must pass before they are truly applied, have received more and more attention in recent years. The technologies related to automatic valet parking and memory parking have also been extensively studied by the industry flied and academia flied. Motion planning is an important part of ensuring that autonomous vehicles can obtain reasonable driving trajectories, achieve driving goals, and avoid collision. Therefore, this paper summarizes the previous research on vehicle motion planning on unstructured environment such as parking area, and proposes a hierarchical vehicle motion planning method for the cursing in parking area. Three levels of planning, including reference trajectory generation based on APF method, initial value extraction with dynamic programming and trajectory optimization with TEB algorithm are combined to obtain a reasonable trajectory for autonomous vehicle in real time. In order to verify the above algorithm, the vehicle experiment is carried out in a parking area in Suzhou, China. Scenes such as cruise and straight-way avoidance are tested to illustrate the effectiveness of the algorithm. The results show that the proposed algorithm can effectively solve these scenes. In the future, more scenes will be tested to verify the effectiveness of the above algorithms.