An Automated Parking Trajectory Planning Method Based on Safe Parking Corridors

For the auto-parking trajectory planning problem, one method is to transform it into an optimal control problem (OCP) and tackle it. However, it is challenging for planning a safe, comfortable trajectory in the open space with various irregular obstacles or narrow scenes due to non-convex obstacle avoidance constraints on a large scale and complete obstacle avoidance in narrow scenes. Inspired by this, a trajectory planning algorithm based on safe parking corridors is proposed. Firstly, an initial trajectory is determined by applying Hybrid A* algorithm, while a precise collision detection module is also introduced to enhance the reliability of Hybrid A* specifically in narrow scenarios. Such obtained trajectory is used as warm start to accelerate the solution of OCP. Additionally, the safe parking corridor constraint is adopted to represent the obstacle avoidance constraint. In this context, not only the computational scale will be reduced, but also the environmental complexity will be avoided. Simulation results demonstrate that the proposed method exhibits robustness with significant improvement in the success rate of planning in narrow and intricate settings.