A Parking Trajectory Planning Method based on Geometric Constraints

Planning a safe trajectory is a crucial procedure for autonomous parking. In existing approaches, parking trajectories arc usually designed without considering the evolution from one configuration to the next configuration. Hence, these trajectories are probably unfeasible in some scenarios. To address this problem, a guaranteed collision-free trajectory planning method is proposed in this paper. A dynamic optimization problem is built, which includes vehicle kinematics, collision avoidance and physical constraints. To describe collision avoidance constraints, the concepts of virtual protection frame (VPF) and magnification parameter are introduced. It is highlighted that a novel collision detection criterion is raised from the view of geometry. In the parallel parking scenario, it is illustrated that the trajectory generated by the proposed method is more feasible than that generated by the conventional method.