Research on Hierarchical Motion Control of Automatic Parking Based on Variable Radius Reeds-Shepp Method

Abstract Automatic parking technology is affected by environmental factors such as narrow space and many obstacles. It often faces problems such as difficult planning of trajectory bending and poor trajectory tracking. In this paper, an automatic parking hierarchical real-time motion control system is designed. In the upper planning controller, a variable radius Reeds-Shepp curve method is proposed. The optimal curvature radius is selected in different parts of the path to reduce unnecessary turns and curve segments, which is more adaptable to complex environments. It can realize efficient and flexible search for vertical/oblique parking and out-of-parking spaces under complex conditions of narrow parking spaces. In the lower tracking controller, the linear quadratic regulation method is used to track the planned parking path. Compared with the traditional path tracking methods such as pure tracking, the tracking accuracy is higher. Finally, the Matlab/Simulink-Prescan co-simulation environment is built. The experimental results show that the proposed variable radius Reeds-Shepp curve method can quickly search out the optimal parking trajectory of entering/leaving the target parking space, and the linear quadratic method can also achieve parking control more accurately.