Automatic parking path re-planning strategy based on optimal control

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), May.2025, Vol.48 No.5: 598-605

DOI: 10.3969/j.issn.1003-5060.2025.05.004

Authors

HU Yanping $ ^{1} $, WANG Yutao $ ^{1} $, ZHAO Linfeng $ ^{2} $, LI Xuchen $ ^{1} $, YIN Haoran $ ^{1} $

(1. School of Mechanical Engineering, Hefei University of Technology, Hefei 230009, China; 2. School of Automobile and Traffic Engineering, Hefei University of Technology, Hefei 230009, China)

Abstract and Keywords

Abstract: The irregular parking in narrow spaces will greatly increase the difficulty of parking. To solve the problem of parking difficulties caused by narrow spaces, this paper proposes a parking path planning method based on optimal control. According to the vehicle posture and parking position, and considering vehicle kinematic constraints, mechanical constraints and parking space boundary constraints, the obstacles in the environment are transformed into convex polygon constraints, the path is planned with the shortest parking time as the performance index, and the variables are completely discretized by orthogonal collocation on finite elements to accelerate the solving process of the optimal control problem, thereby meeting the real-time requirements of parking. Considering the impact of parking space detection errors and road conditions on actual parking, a path re-planning strategy is proposed to further improve the parking success rate. Simulations and real vehicle experiments show that the proposed method can complete the parking action quickly without collision in a narrow parking space.

Keywords: automatic parking; path planning; optimal control; narrow space; real vehicle experiment

作者信息

胡延平 $ ^{1} $，汪彧涛 $ ^{1} $，赵林峰 $ ^{2} $，李旭辰 $ ^{1} $，尹浩然 $ ^{1} $

(1. 合肥工业大学机械工程学院，安徽合肥 230009；2. 合肥工业大学汽车与交通工程学院，安徽合肥 230009)

摘要和关键词

摘要: 车位狭窄时停车不规范会大大增加泊车难度。针对狭窄空间带来的泊车困难问题,文章提出一种基于最优控制的泊车路径规划方法。根据自车位姿和车位位置,考虑车辆运动学约束、机械约束、泊车空间边界约束,将环境中的障碍物转化为凸多边形约束,以最短泊车时间为性能指标规划路径,并通过正交配置有限元法对变量进行完全离散,加速最优控制问题求解过程,以满足泊车实时性要求;考虑到车位检测误差和路面状况对实际泊车的影响,为进一步提高泊车成功率,提出路径重规划策略。仿真和实车试验结果表明,该文所提的自动泊车路径规划方法能够在狭窄泊车空间内无碰撞地快速完成泊车动作。

关键词: 自动泊车；路径规划；最优控制；狭窄空间；实车试验

Funding

安徽省科技重大专项资助项目（202103a05020018）

合肥工业大学学报(自然科学版)

Navigation