A proactive control-based dynamic allocation model for high-density autonomous parking lots
Article Text (Baidu Translation)
-
摘要: 为解决智能化和集约化停车设施发展下高密度自动驾驶停车场车位动态分配优化问题，提出主动式前馈预测控制的车位动态分配模型(P3DD)；基于模型预测控制框架，针对停车场内车辆到达、离开以及潜在的冲突进行了车位分配的动态优化决策；构建了卷积神经网络-长短期记忆网络(CNN-LSTM)以滚动预测未来车辆到达和离开时间；以最小化综合停车成本为目标，构建了综合考虑行驶距离、移动次数、取车等待时间和响应失败率的优化目标函数，在预测控制环境中采用禁忌搜索的优化算法优化实时分配决策；基于杭州市多个停车场的实际数据，在不同停车布局和停车需求与容量比条件下，测试了P3DD优化性能。研究结果表明：相较于采用被动启发式规则的基线模型，P3DD能够通过主动预测和实时优化决策，直接面向停车系统性能指标进行优化，在9个算例中改进率范围为44.8%~56.5%，车辆综合停车成本平均下降48.0%；CNN-LSTM模型对未来2 h车辆到达/离开时间的预测准确度平均达0.81；在不同停车布局和停车需求与容量比条件下的测试显示，随着停车布局堆栈深度增加，P3DD的优化效果提升，在停车需求与供给均衡时优化效果最佳；此外，P3DD具备良好的适应性和扩展性，为高密度停车场的动态资源分配提供了一种高效且灵活的解决方案。Abstract: A proactive predictive dynamic parking space allocation model (P3DD) was proposed to address the dynamic parking space allocation optimization problem in high-density automated parking lots amid the development of intelligent and intensive parking facilities. Based on the model predictive control (MPC) framework, dynamic optimization decisions for parking space allocation were made by considering the vehicle arrival, departure, and potential conflicts within the parking lot. Meanwhile, a convolutional neural network-long short-term memory (CNN-LSTM) framework is constructed for the rolling prediction of future vehicle arrival and departure time. An optimization objective function that comprehensively considers driving distance, vehicle movement times, waiting time for vehicle retrieval, and response failure rates was developed to minimize the comprehensive parking cost. A tabu search-based optimization algorithm was employed in the predictive control environment to optimize real-time allocation decisions. Based on real-world data from multiple parking lots in Hangzhou, the P3DD model's optimization performance was tested under various parking layouts and parking demand-to-capacity ratios. The results indicate that compared to the baseline model adopting reactive heuristic rules, the P3DD model can directly optimize the parking system performance indicators via proactive prediction and real-time optimization decisions, with improvement rates ranging from 44.8% to 56.5% across nine test cases, and the average comprehensive parking cost decreasing by 48.0%. The CNN-LSTM model achieves an average prediction accuracy of 0.81 for vehicle arrival/departure within the next two hours. Tests under different parking layouts and demand-to-capacity ratios demonstrate that as the stack depth of parking layouts increases, the optimization effect of the P3DD model improves, with optimal performance under balanced parking demand and supply. Furthermore, the P3DD model demonstrates sound adaptability and scalability, providing an efficient and flexible solution for dynamic resource allocation in high-density parking lots.
-
表 1 算例测试结果
Table 1. Results of simulation experiments
算例编号 1 2 3 4 5 6 7 8 9 基线成本/(s·veh-1) 141.0 148.8 153.9 151.1 152.0 157.7 153.6 161.2 147.9 优化成本/(s·veh-1) 61.4 67.3 82.4 82.4 83.9 85.1 84.2 80.3 78.6 改进率/% 56.5 54.8 46.5 45.4 44.8 46.0 45.2 50.2 46.8 预测准确度 0.84 0.82 0.85 0.81 0.80 0.84 0.79 0.80 0.79 响应时间/s 1.59 1.67 1.65 1.66 1.64 1.75 1.62 1.64 1.66 
下载: 导出CSV
-
[1] 李瑞敏, 戴晶辰. 自动驾驶影响下的出行行为研究综述[J]. 交通运输工程学报, 2022, 22(3): 41-54. doi: 10.19818/j.cnki.1671-1637.2022.03.003 LI Rui-min, DAI Jing-chen. Review on impact of autonomous driving on travel behaviors[J]. Journal of Traffic and Trans-portation Engineering, 2022, 22(3): 41-54. doi: 10.19818/j.cnki.1671-1637.2022.03.003 [2] 胡笳, 罗书源, 赖金涛, 等. 自动驾驶对交通运输系统规划的影响综述[J]. 交通运输系统工程与信息, 2021, 21(5): 52-65, 76.HU Jia, LUO Shu-yuan, LAI Jin-tao, et al. A review of the impact of autonomous driving on transportation planning[J]. Journal of Transportation Systems Engineering and Infor-mation Technology, 2021, 21(5): 52-65, 76. [3] BANZHAF H, NIENHüSER D, KNOOP S, et al. The future of parking: A survey on automated valet parking with an outlook on high density parking[C]//IEEE. 2017 IEEE Intelligent Vehicles Symposium (Ⅳ). New York: IEEE, 2017: 1827-1834. [4] JANG C, KIM C, LEE S, et al. Re-plannable automated parking system with a standalone around view monitor for narrow parking lots[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(2): 777-790. doi: 10.1109/TITS.2019.2891665 [5] MENG Q H, QIAN C J, SUN Z Y, et al. Autonomous parking method based on improved A* algorithm and model predictive control[J]. Nonlinear Dynamics, 2025, 113(7): 6839-6862. doi: 10.1007/s11071-024-10456-7 [6] TIMPNER J, FRIEDRICHS S, VAN BALEN J, et al. K-Stacks: High-density valet parking for automated vehicles[C]//IEEE. 2015 IEEE Intelligent Vehicles Symposium (Ⅳ). New York: IEEE, 2015: 895-900. [7] BANZHAF H, QUEDENFELD F, NIENHÜSER D, et al. High density valet parking using k-deques in driveways[C]//IEEE. 2017 IEEE Intelligent Vehicles Symposium (Ⅳ). New York: IEEE, 2017: 1413-1420. [8] NOURINEJAD M, BAHRAMI S, ROORDA M J. Designing parking facilities for autonomous vehicles[J]. Transportation Research Part B: Methodological, 2018, 109: 110-127. doi: 10.1016/j.trb.2017.12.017 [9] 于红媚, 叶晓飞, 汪义路, 等. 基于NSGA-Ⅱ算法的自动驾驶停车泊位优化模型[J]. 宁波大学学报(理工版), 2024, 37(3): 57-69.YU Hong-mei, YE Xiao-fei, WANG Yi-lu, et al. Optimizing model for automatic parking based on NSGA-Ⅱ algorithm[J]. Journal of Ningbo University (Natural Science & Engi-neering Edition), 2024, 37(3): 57-69. [10] SIDDIQUE P J, GUE K R, USHER J S. Puzzle-based parking[J]. Transportation Research Part C: Emerging Techno-logies, 2021, 127: 103112. doi: 10.1016/j.trc.2021.103112 [11] FERREIRA M, DAMAS L, CONCEICAO H, et al. Self-automated parking lots for autonomous vehicles based on vehicular AD HOC networking[C]//IEEE. 2014 IEEE Intelligent Vehicles Symposium Proceedings. New York: IEEE, 2014: 472-479. [12] NUNES R, MOREIRA-MATIAS L, FERREIRA M. Using exit time predictions to optimize self automated parking lots[C]//IEEE. 17th International IEEE Conference on Intelligent Transportation Systems (ITSC). New York: IEEE, 2014: 302-307. [13] D'OREY P M, AZEVEDO J, FERREIRA M. Exploring the solution space of self-automated parking lots: An empirical evaluation of vehicle control strategies[C]//IEEE. 2016 IEEE 19th International Conference on Intelligent Trans-portation Systems (ITSC). New York: IEEE, 2016: 1134-1140. [14] HOU J, CHEN G, LI Z J, et al. Hybrid residual multiex-pert reinforcement learning for spatial scheduling of high-density parking lots[J]. IEEE Transactions on Cybernetics, 2024, 54(5): 2771-2783. doi: 10.1109/TCYB.2023.3312647 [15] GU Z Y, NAJMI A, SABERI M, et al. Macroscopic parking dynamics modeling and optimal real-time pricing considering cruising-for-parking[J]. Transportation Research Part C: Emerging Technologies, 2020, 118: 102714. doi: 10.1016/j.trc.2020.102714 [16] DUDAKL N, BAYKASOGLU A. A simulation-optimization-based planning and control system for operations of fully automated parking systems[J]. Computers & Industrial Engineering, 2024, 189: 109977. [17] HEGYI A, DE SCHUTTER B, HELLENDOORN J. Optimal coordination of variable speed limits to suppress shock waves[J]. IEEE Transactions on Intelligent Transportation Systems, 2005, 6(1): 102-112. doi: 10.1109/TITS.2004.842408 [18] HAN Y, YU H, LI Z B, et al. An optimal control-based vehicle speed guidance strategy to improve traffic safety and efficiency against freeway jam waves[J]. Accident Analysis & Prevention, 2021, 163: 106429. [19] ABOUDOLAS K, PAPAGEORGIOU M, KOUVELAS A, et al. A rolling-horizon quadratic-programming approach to the signal control problem in large-scale congested urban road networks[J]. Transportation Research Part C: Emerging Technologies, 2010, 18(5): 680-694. doi: 10.1016/j.trc.2009.06.003 [20] LIN S, DE SCHUTTER B, XI Y G, et al. Fast model predictive control for urban road networks via MILP[J]. IEEE Transactions on Intelligent Transportation Systems, 2011, 12(3): 846-856. doi: 10.1109/TITS.2011.2114652 [21] FENG N X, ZHANG F, LIN J Z, et al. Statistical analysis and prediction of parking behavior[C]//Springer. Network and Parallel Computing: 16th IFIP WG 10.3 International Conference. Berlin: Springer, 2019: 93-104. [22] XU M S, GAO S, ZHENG J Y, et al. Day-ahead electric vehicle charging behavior forecasting and schedulable capacity calculation for electric vehicle parking lot[J]. Energy, 2024, 309: 133090. doi: 10.1016/j.energy.2024.133090 [23] ZHANG F, FENG N X, LIU Y N, et al. PewLSTM: Periodic LSTM with weather-aware gating mechanism for parking behavior prediction[C]//IJCAI. Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelli-gence. Yokohama: IJCAI, 2020: 4424-4430. [24] ZHANG F, LIU Y N, FENG N X, et al. Periodic weather-aware LSTM with event mechanism for parking behavior prediction[J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(12): 5896-5909. doi: 10.1109/TKDE.2021.3070202 [25] ZOU M, WANG Q, LIU S A. Optimization of parking space allocation for automated parking system of paternoster type by genetic algorithm[C]//CCDC. 2019 Chinese Control And Decision Conference. Nanchang: CCDC, 2019: 3834-3838. [26] BAHRAMI S, ROORDA M. Autonomous vehicle parking policies: A case study of the city of Toronto[J]. Trans-portation Research Part A: Policy and Practice, 2022, 155: 283-296. doi: 10.1016/j.tra.2021.11.003 [27] 冯驰, 梅振宇. 基于自动驾驶车辆调度的停车系统收费策略优化[J]. 浙江大学学报(工学版), 2024, 58(1): 87-95.FENG Chi, MEI Zhen-yu. Optimization of parking charge strategy based on dispatching autonomous vehicles[J]. Journal of Zhejiang University (Engineering Science), 2024, 58(1): 87-95. [28] 宋现敏, 湛天舒, 李海涛, 等. 考虑用户成本和泊位利用率的停车预约分配模型[J]. 吉林大学学报(工学版), 2025, 55(4): 1287-1297.SONG Xian-min, ZHAN Tian-shu, LI Hai-tao, et al. Reser-vation and allocation model considering user cost and utilization of parking space[J]. Journal of Jilin University (Engineering and Technology Edition), 2025, 55(4): 1287-1297. [29] MA C X, HUANG X T, LI J C. A review of research on urban parking prediction[J]. Journal of Traffic and Trans-portation Engineering: English Edition, 2024, 11(4): 700-720. [30] LING T W, ZHU X, ZHOU X L, et al. ParkLSTM: Periodic parking behavior prediction based on LSTM with multi-source data for contract parking spaces[C]//Springer. Wireless Algorithms, Systems, and Applications: 17th International Conference. Berlin: Springer, 2021: 262-274. [31] 李剑峰. 单点信号配时改进方法研究[D]. 长春: 吉林大学, 2009.LI Jian-feng. Study on improved methods of signal timing at isolated intersection[D]. Changchun: Jilin University, 2009. -
图(5) / 表(1)
计量
- 文章访问数: 109
- HTML全文浏览量: 50
- PDF下载量: 13
- 被引次数: 0
