摘要详情

ID / 提交时间

106 / 2025-05-15 11:06:04

标题

Path Planning for Automated Valet Parking in Underground Parking Lots Based on UWB and Improved Bidirectional A* Algorithm

关键字

automated valet parking, path planning, A*, UWB

主题及专题

Special Sessions > SSL 13 Advancements in Intelligent Fault Prediction, Diagnosis, and Sustainable Maintenance of Vehicle Critical Systems

状态

全文待审

作者

Liu Run / SHENZHEN TECHNOLOGY UNIVERSITY

Heyan Li / SHENZHEN TECHNOLOGY UNIVERSITY

Wenhao Zhang / SHENZHEN TECHNOLOGY UNIVERSITY

Haoyun Yuan / SHENZHEN TECHNOLOGY UNIVERSITY

Junde Li / SHENZHEN TECHNOLOGY UNIVERSITY

Jiantong Wu / SHENZHEN TECHNOLOGY UNIVERSITY

摘要

In increasingly complex underground parking environments, traditional path planning algorithms often struggle with low efficiency and poor path stability when handling long-distance parking tasks. To address these challenges, this paper proposes an improved bidirectional A* path planning method guided by virtual Ultra-Wideband (UWB) anchor points. The approach enhances the heuristic cost function by incorporating information from both forward and reverse search frontiers, which accelerates path convergence and reduces redundant node expansion. Additionally, virtual relay points are introduced via UWB positioning systems to guide the path direction, and the global task is executed in multiple stages to further improve search efficiency and robustness. Simulation results in the university's underground parking lot demonstrate that, compared with traditional A* and bidirectional A* algorithms, the proposed method significantly reduces both search time and node expansion, particularly in long-distance and multi-intersection scenarios. These results validate the feasibility and practicality of the proposed method for large-scale indoor automated valet parking (AVP) applications.