A Feasible Parking Algorithm in Form of Path Planning and Following

Automatic parking technology is the research trend in auto area in recent years. The parking algorithm is usually decomposed into two parts: path planning with geometrical method and path following. Actually, many papers have contributed on improving the performance of automatic parking algorithm. Among the existing algorithms, the jerks of curvature at some reference path points are hard to be avoided. The problem increases the difficulty of tracking. In path tracking, the simple proportion-integration-differential(PID) controller can not meet the requirements on high precision and stability. Therefore, this paper aims at generating smooth trajectories and tracking them precisely and stably. Firstly, the proposed algorithm utilizes circular arcs to obtain initial reference parking paths in view of passable constraints like slots' boundaries and road boundaries. Secondly, the algorithm chooses appropriate Bezier curves to ease the jerks at path joints. Finally, the controller based on model predictive control(MPC) is designed for path tracking. The automatic parking algorithm is verified in MATLAB and vehicle dynamics simulation software PreScan. Simulation results show that the proposed algorithm is feasible and efficient.