Parallel Parking Path Planning in Narrow Space Based on a Three-Stage Curve Interpolation Method

A parallel parking path planning method aiming at the automatic parking application scene of parallel parking in narrow space of urban roadside parking lot is proposed in the paper, based on a three-stage curve interpolation method. After the vehicle finds and detects the idle parking space through the environment perception technology, the parking path is constructed using the clothoid curve, circular arc, and fifth-degree polynomial curve by performing reverse analysis starting from the parking termination point. Considering the constraints of vehicle technical parameters and collisions avoidance during the parking process, a feasible starting area for parking is established, and the constrained optimization problem is solved to generate a path curve that satisfies curvature continuity and parking constraints. Three different parking conditions are proposed for parking simulation experiments, and the path planning method proposed in the paper (method I), the circular arc and fifth-degree polynomial curve path planning method (method II), and the fifth-degree polynomial curve path planning method (method III) are compared and analyzed. The results show that the path planning method proposed in the paper can effectively eliminate the sudden change point of curvature, constrain the path curvature and ensure the collision-free requirement for obstacles when entering the parking space for different parking spaces or different starting parking points.