UGV Parking Planning Based on Swarm Optimization and Improved CBS in High-Density Scenarios for Innovative Urban Mobility
Author:
Zeng Dequan123, Chen Haotian1, Yu Yinquan1, Hu Yiming12, Deng Zhenwen34ORCID, Leng Bo23, Xiong Lu23, Sun Zhipeng2
Affiliation:
1. School of Mechanical Electronic and Vehicle Engineering, East China Jiaotong University, Nanchang 330013, China 2. Nanchang Automotive Institution of Intelligence & New Energy, Nanchang 330052, China 3. School of Automotive Studies, Tongji University, Shanghai 201804, China 4. Institute of Computer Application Technology, NORINCO Group, Beijing 100089, China
Abstract
The existence of information silos between vehicles and parking lots means that Unmanned Ground Vehicles (UGVs) repeatedly drive to seek available parking slots, resulting in wasted resources, time consumption and traffic congestion, especially in high-density parking scenarios. To address this problem, a novel UGV parking planning method is proposed in this paper, which consists of cooperative path planning, conflict resolution strategy, and optimal parking slot allocation, intending to avoid ineffective parking seeking by vehicles and releasing urban traffic pressure. Firstly, the parking lot induction model was established and the IACA–IA was developed for optimal parking allocation. The IACA–IA was generated using the improved ant colony algorithm (IACA) and immunity algorithm. Compared with the first-come-first-served algorithm (FCFS), the normal ant colony algorithm (NACA), and the immunity algorithm (IA), the IACA–IA was able to allocate optimal slots at a lower cost and in less time in complex scenarios with multi-entrance parking lots. Secondly, an improved conflict-based search algorithm (ICBS) was designed to efficiently resolve the conflict of simultaneous path planning for UGVs. The dual-layer objective expansion strategy is the core of the ICBS, which takes the total path cost of UGVs in the extended constraint tree as the first layer objective, and the optimal driving characteristics of a single UGV as the second layer objective. Finally, three kinds of load-balancing and unbalanced parking scenarios were constructed to test the proposed method, and the performance of the algorithm was demonstrated from three aspects, including computation, quality and timeliness. The results show that the proposed method requires less computation, has higher path quality, and is less time-consuming in high-density scenarios, which provide a reasonable and efficient solution for innovative urban mobility.
Subject
Artificial Intelligence,Computer Science Applications,Aerospace Engineering,Information Systems,Control and Systems Engineering
Reference33 articles.
1. Liu, Q., Li, Z., Yuan, S., Zhu, Y., and Li, X. (2021). Review on Vehicle Detection Technology for Unmanned Ground Vehicles. Sensors, 21. 2. Bi-level objective berth allocation model based on the minimum total system time consumption;Yang;Univ. Shanghai Sci. Technol.,2021 3. Zhao, Z., Zhang, Y., Shi, J., Long, L., and Lu, Z. (2022). Robust Lidar-Inertial Odometry with Ground Condition Perception and Optimization Algorithm for UGV. Sensors, 22. 4. Research on Optimal Setting of Shared parking Space adjacent to private parking lot in urban center Area;Ji;Syst. Eng. Theory Pract.,2020 5. Research on Automatic Parking Systems Based on Parking Scene Recognition;Ma;IEEE Access,2017
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|