Maneuver Planning for Automatic Parking with Safe Travel Corridors: A Numerical Optimal Control Approach

Abstract

An optimal control problem is a straightforward description of a generic automatic parking maneuver planning scheme. However, the dimensionality of the optimal control problem is usually high, primarily because of the large-scale collision avoidance constraints. Among the complete collision avoidance constraints, some parts are redundant because the ego vehicle may not be subject to collisions with all of the obstacles at every moment. An idea to simplify the collision avoidance constraints is to plan a coarse path/trajectory, and then to construct a surrounding corridor which naturally separates the ego vehicle from the obstacles. This idea is well known as safe flight corridors (SFCs) in the research area of unmanned aerial vehicle (UAV) path planning. But SFCs are not directly applicable to automatic parking cases because the ego vehicle cannot be taken as a mass point or a circle in the tiny parking scenarios. To make the SFCs work for automatic parking schemes, we develop a concept called Safe Travel Corridors (STCs), which requires different parts of the ego vehicle to stay in different safe corridors. Through this, a reduced-scale optimal control problem is formulated, and the problem scale is completely irrelevant to the complexity of the environment. The efficiency and robustness of the proposed maneuver planner are evaluated through 96 benchmark cases.