基于分层架构的自主代客泊车路径规划  

Autonomous valet parking path planning based on hierarchical architecture

作　　者：张家旭 王志伟 郭崇[1] 赵健[1] 周时莹 Zhang Jiaxu;Wang Zhiwei;Guo Chong;Zhao Jian;Zhou Shiying(State Key Laboratory of Automotive Simulation and Control,Jilin University,Changchun 130022,China;Intelligent Network R&D Institute,China FAW Group Co.,Ltd.,Changchun 130011,China)

机构地区：[1]吉林大学汽车仿真与控制国家重点实验室,长春130022 [2]中国第一汽车集团有限公司智能网联研发院,长春130011

出　　处：《东南大学学报（自然科学版）》2021年第5期883-888,共6页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(51775235).

摘　　要：为了解决最后一公里自动驾驶问题,提出一种基于分层架构的自主代客泊车路径规划算法.首先,利用栅格扫描算法快速准确地将自主代客泊车环境地图转化成为Voronoi图,量化自主代客泊车环境中任意栅格格网区域与其最近障碍物距离.随后,利用A*算法规划出全局自主代客泊车路径,并采用优先队列数据实现A*算法的开放列表来提升其计算效率.最后,基于自主代客泊车环境Voronoi图实现汽车碰撞检测,并采用改进动态窗口法沿着全局自主代客泊车路径规划出满足汽车非完整约束和机械约束的无碰撞路径,扩展传统动态窗口法的可行解空间和降低其保守性.在VC++6.0环境中验证所提出的自主代客泊车路径规划算法的可行性和有效性,结果表明,该算法可以安全、快速地引导汽车到达目标泊车位附近,为汽车后续执行泊车操作奠定基础.In order to solve the problem of automatic driving in the last kilometer,a novel autonomous valet parking path planning algorithm is proposed based on the hierarchical architecture.Firstly,the raster scanned algorithm is used to quickly and accurately convert the map of autonomous valet parking environment into Voronoi diagram,which can quantify the distance between any grid area and the nearest obstacle in the environment.Then,the A*algorithm is used to plan the global autonomous valet parking path,and the priority queue is used to realize the open list of A*algorithm to improve its computational efficiency.Finally,the Voronoi diagram in the autonomous valet parking environment is used to detect vehicle collision,and along the global autonomous valet parking path,a collision free path satisfying the non-holonomic and mechanical constraints of vehicle is planned by the improved dynamic window method,which expands the feasible solution space of the traditional dynamic window method and reduces its conservatism.The feasibility and effectiveness of the proposed autonomous valet parking path planning algorithm are verified in VC++6.0 environment.The results show that the proposed algorithm can safely and quickly guide the vehicle to the near target parking space,and lay the foundation for the car to perform subsequent parking operations.

关 键 词：栅格扫描算法 自主代客泊车 路径规划 A*算法 改进动态窗口法 

分 类 号：U461.1[机械工程—车辆工程]