下载掌阅APP，畅读海量书库
立即打开
Li Chao, Du Jianyu, Li Jianfeng, Wang Yinan, Liu Bin, Lü Ying
FAW Group Corporation-Automated & Connected Development Institute
Abstract: Path planning algorithms directly affect the performance of the automatic parking assist system(APA).RS algorithm is very suitable for embedded code transplantation and can quickly generate the shortest parking path due to its low computational complexity, so it is widely used in the parking path planning algorithms.However, because the RS algorithm does not consider the constraint of curvature mutation, the phenomenon of turning the wheel on the spot often occurs during the actual parking process, which will not only accelerate tire wear, but also affect the driver's driving experience.In order to solve the curvature instantaneous problem of the RS algorithm, this paper proposes a continuous curvature parking path planning algorithm, that is, to optimize and upgrade the RS algorithm with clothoid and straight line, and change the connection of arcs and straight lines used by the RS algorithm to clothoid arcs and straight lines, and so on the basis of retaining the advantages of RS algorithm, it can effectively solve the problem of curvature mutation problem during parking process.Finally, the effectiveness of the proposed algorithm is tested and verified by simulation and real vehicle experiments in two common parking scenarios.The experimental results show that the proposed algorithm is not only simple for embedded transplantation, but also can effectively solve the curvature mutation problem during parking process.
Key words: RS, APA, continuous-curvature, path planner