都市化與車輛數量的持續成長使得路邊停車資源長期供不應求，駕駛人往往需要花費大量時間在目的地周邊進行無效繞行，造成交通壅塞、能源浪費與碳排放增加。現有研究雖已針對停車位可用性預測提出多種深度學習模型，但多數方法仍難以有效結合預測結果與決策策略，亦未能處理多車輛同時競逐停車資源的動態互動情境。基於此，本研究提出一套整合停車預測與智慧導引決策之多智能體強化學習框架，以協助駕駛在最短時間內找到可用停車位。
在預測層面，本研究採用多視角時空圖卷積網路（MV-STGCN）建模停車可用率於時間與空間維度上的相關性，並將原始MV-STGCN中的Global GAT替換為改良後的DPRGAT模組，以結合距離感知的Parking Rank（DPR）評分機制，捕捉駕駛目的地導向的行為偏好，使預測結果更貼近真實決策情境。於決策層面，本研究採用SEEDRL架構建構多智能體強化學習環境，模擬多輛車輛在競爭式停車環境中的互動，並透過集中訓練、分散執行（CTDE）方式學習可即時適應環境變動的停車搜尋策略。
模擬結果顯示，所提出之整合架構在尖峰與離峰時段皆能顯著縮短平均停車搜尋時間，並使最終停車位置更接近目的地。完整系統（MV-STGCN + DPRGAT + SEEDRL）亦在訓練效率與策略穩定性方面展現最佳表現。實驗驗證本研究提出之距離感知停車預測與多智能體決策整合架構，能有效提升智慧城市環境中的停車搜尋效率，降低駕駛繞行成本，並具備良好的可擴展性與實務應用潛力。

This study proposes an integrated framework for intelligent on-street parking guidance that jointly addresses parking availability prediction and real-time decision-making under multi-agent competition. A multi-view spatial–temporal graph convolutional network (MV-STGCN) is employed to model temporal dynamics and spatial correlations of parking utilization. To incorporate destination-oriented driver preferences, the global attention module is replaced with a Distance-aware Parking Rank Graph Attention Network (DPRGAT), which embeds a distance-aware service capability score into the spatial representation. For decision-making, a scalable multi-agent reinforcement learning architecture based on SEEDRL is developed to learn efficient parking search strategies through centralized training and decentralized execution.
Experimental evaluations demonstrate that the proposed framework significantly reduces parking search time and yields parking positions closer to drivers’ destinations compared with baseline reinforcement learning models. The full system also exhibits improved training stability and computational efficiency, indicating strong potential for deployment in large-scale smart city applications.

[1] 王彥嵐, "於霧計算架構下以LSTM模型預測空餘路邊停車位," 國立政治大學, 2019. [Online]. Available: https://doi.org/10.6814%2fNCCU201900455
[2] 交通部運輸研究所, "2022年臺灣公路容量手冊," 2022.
[3] 臺北市政府 (2022). 1110628-臺北市文山區都市計畫主要計畫(第二次通盤檢討)案(第一階段)-計畫書.
[4] 臺北市政府工務局. "交通流量調查資料." https://bote.gov.taipei/cp.aspx?n=E0C93DC334AE8028&s=D8FF1127E9754399 (accessed.
[5] 臺北市資料大平臺 (2024). 臺北市停車位數時間數列統計資料. [Online] Available: https://data.taipei/dataset/detail?id=301a7699-75eb-4dff-8f3f-362984b0581c
[6] 臺北市資料大平臺 (2024). 臺北市機動車輛登記數. [Online] Available: https://data.taipei/dataset/detail?id=ce44d079-6588-4dad-9232-9ecb88f205ff
[7] 臺北市停車管理工程處 (2023). 112年度臺北市汽機車停車供需調查. [Online] Available: https://pma.gov.taipei/News.aspx?n=65D85809A5ABC2C9&sms=6AAEA2F8F4E6DFD2
[8] 數位發展部數位產業署智慧城鄉生活應用發展計畫專案委辦. "停車服務一條龍 找位管理更從容." https://www.twsmartcity.org.tw/smart_application/13 (accessed 7/4, 2025).
[9] J. Arjona, M. P. Linares, J. Casanovas-Garcia, and J. J. Vázquez, "Improving Parking Availability Information Using Deep Learning Techniques," Transportation Research Procedia, vol. 47, pp. 385-392, 2020/01/01/ 2020, doi: https://doi.org/10.1016/j.trpro.2020.03.113.
[10] J. Arjona, M. P. Linares, and J. Casanovas, "A deep learning approach to real-time parking availability prediction for smart cities," presented at the Proceedings of the Second International Conference on Data Science, E-Learning and Information Systems, Dubai, United Arab Emirates, 2019. [Online]. Available: https://doi.org/10.1145/3368691.3368707.
[11] A. F. Auer, Shelley; Lockwood, Stephen, "History of Intelligent Transportation Systems," U.S. Department of Transportation, ITS Joint Program Office, 2016. Accessed: 2023-06-16. [Online]. Available: chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.naco.org/sites/default/files/documents/01_Smith.pdf
[12] E. AvŞAr, Y. C. Anar, and A. Ö. Polat, "Parking lot occupancy prediction using long short-term memory and statistical methods," MANAS Journal of Engineering, vol. 10, no. 1, pp. 35-41, 2022, doi: 10.51354/mjen.986631.
[13] K. S. Awaisi, A. Abbas, H. A. Khattak, A. Ahmad, M. Ali, and A. Khalid, "Deep reinforcement learning approach towards a smart parking architecture," Cluster Computing, vol. 26, no. 1, pp. 255-266, 2022, doi: 10.1007/s10586-022-03599-y.
[14] F. M. Awan, Y. Saleem, R. Minerva, and N. Crespi, "A Comparative Analysis of Machine/Deep Learning Models for Parking Space Availability Prediction," Sensors, vol. 20, no. 1, p. 322, 2020. [Online]. Available: https://www.mdpi.com/1424-8220/20/1/322.
[15] D. Ayala, O. Wolfson, B. Xu, B. Dasgupta, and J. Lin, "Parking slot assignment games," presented at the Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, Chicago, Illinois, 2011. [Online]. Available: https://doi.org/10.1145/2093973.2094014.
[16] C. Badii, P. Nesi, and I. Paoli, "Predicting Available Parking Slots on Critical and Regular Services by Exploiting a Range of Open Data," IEEE Access, vol. 6, pp. 44059-44071, 2018, doi: 10.1109/ACCESS.2018.2864157.
[17] F. Bock, S. D. Martino, and A. Origlia, "A 2-Step Approach to Improve Data-driven Parking Availability Predictions," presented at the Proceedings of the 10th ACM SIGSPATIAL Workshop on Computational Transportation Science, Redondo Beach, CA, USA, 2017. [Online]. Available: https://doi.org/10.1145/3151547.3151550.
[18] S. S. Channamallu, S. Kermanshachi, J. M. Rosenberger, and A. Pamidimukkala, "Parking occupancy prediction and analysis - a comprehensive study," Transportation Research Procedia, vol. 73, pp. 297-304, 2023/01/01/ 2023, doi: https://doi.org/10.1016/j.trpro.2023.11.921.
[19] S. Cheung and P. Varaiya, "Traffic Surveillance by Wireless Sensor Networks: Final Report," 01/01 2007.
[20] K. Cho, B. van Merriënboer, D. Bahdanau, and Y. Bengio, "On the Properties of Neural Machine Translation: Encoder–Decoder Approaches," Doha, Qatar, October 2014: Association for Computational Linguistics, in Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, pp. 103-111, doi: 10.3115/v1/W14-4012. [Online]. Available: https://aclanthology.org/W14-4012/https://doi.org/10.3115/v1/W14-4012
[21] Z. Deng and D. Wang, "Research on Parking Path Planing Based on A-Star Algorithm," Journal of New Media, vol. 5, no. 1, 2023, doi: 10.32604/jnm.2023.040252.
[22] S. Dong, M. Chen, L. Peng, and H. Li, "Parking rank: A novel method of parking lots sorting and recommendation based on public information," in 2018 IEEE International Conference on Industrial Technology (ICIT), 20-22 Feb. 2018 2018, pp. 1381-1386, doi: 10.1109/ICIT.2018.8352381.
[23] L. Espeholt et al., "IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures," presented at the Proceedings of the 35th International Conference on Machine Learning, Proceedings of Machine Learning Research, 2018. [Online]. Available: https://proceedings.mlr.press/v80/espeholt18a.html.
[24] N. Feng, F. Zhang, J. Lin, J. Zhai, and X. Du, "Statistical Analysis and Prediction of Parking Behavior," in Network and Parallel Computing, Cham, X. Tang, Q. Chen, P. Bose, W. Zheng, and J.-L. Gaudiot, Eds., 2019// 2019: Springer International Publishing, pp. 93-104.
[25] S. S. Ghosal, A. Bani, A. Amrouss, and I. E. Hallaoui, "A Deep Learning Approach to Predict Parking Occupancy using Cluster Augmented Learning Method," in 2019 International Conference on Data Mining Workshops (ICDMW), 8-11 Nov. 2019 2019, pp. 581-586, doi: 10.1109/ICDMW.2019.00088.
[26] A. Goeckner, Y. Sui, N. Martinet, X. Li, and Q. Zhu, "Graph Neural Network‑based Multi‑agent Reinforcement Learning for Resilient Distributed Coordination of Multi‑Robot Systems," presented at the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024. [Online]. Available: https://doi.org/10.1109/IROS58592.2024.10802510.
[27] A. N. Hamidou Soumana, M. Ben Salah, S. Idbraim, and A. Boulouz, "Machine Learning Models in the large-scale prediction of parking space availability for sustainable cities," EAI Endorsed Transactions on Internet of Things, vol. 10, 2023, doi: 10.4108/eetiot.2269.
[28] H. Han et al., "STGCN: A Spatial-Temporal Aware Graph Learning Method for POI Recommendation," in 2020 IEEE International Conference on Data Mining (ICDM), 17-20 Nov. 2020 2020, pp. 1052-1057, doi: 10.1109/ICDM50108.2020.00124.
[29] S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997, doi: 10.1162/neco.1997.9.8.1735.
[30] C. Hongyang, S. Zhang, J. Xie, H. Liu, and G. Chen, "MV-STGCN: Multi-view Spatial-Temporal Graph Convolutional Networks for On-Street Parking Availability Prediction," ACM Trans. Spatial Algorithms Syst., vol. 11, no. 3, p. Article 13, 2025, doi: 10.1145/3732286.
[31] International Organization of Motor Vehicle Manufacturers, "WORLD MOTOR VEHICLE PRODUCTION BY COUNTRY/REGION AND TYPE," 2024, vol. 2025. [Online]. Available: https://www.oica.net/wp-content/uploads/By-country-region-2024.pdf
[32] A. Jose and V. V, "A Stacked Long Short-Term Memory Neural Networks for Parking Occupancy Rate Prediction," in 2021 10th IEEE International Conference on Communication Systems and Network Technologies (CSNT), 18-19 June 2021 2021, pp. 522-525, doi: 10.1109/CSNT51715.2021.9509621.
[33] A. Khanna and R. Anand, "IoT based smart parking system," in 2016 International Conference on Internet of Things and Applications (IOTA), 22-24 Jan. 2016 2016, pp. 266-270, doi: 10.1109/IOTA.2016.7562735.
[34] A. O. Kotb, Y. c. Shen, and Y. Huang, "Smart Parking Guidance, Monitoring and Reservations: A Review," IEEE Intelligent Transportation Systems Magazine, vol. 9, no. 2, pp. 6-16, 2017, doi: 10.1109/MITS.2017.2666586.
[35] H. Kuang, K. Deng, Q. Wang, W. Ye, H. Qu, and J. Li, "Deep meta-learning approach for regional parking occupancy prediction considering heterogeneous and real-time information," Advanced Engineering Informatics, vol. 64, p. 102969, 2025/03/01/ 2025, doi: https://doi.org/10.1016/j.aei.2024.102969.
[36] C.-Y. Lee, "Study on Q-Learning and Deep Q-Learning in Urban Roadside Parking Space Search," Master Academic Thesis, Department of Computer Science, National Chengchi University, 2022. [Online]. Available: https://hdl.handle.net/11296/2j9v22
[37] J.-S. Leu and Z.-Y. Zhu, "Regression-based parking space availability prediction for the Ubike system," IET Intelligent Transport Systems, vol. 9, no. 3, pp. 323-332, 2015/04/01 2015, doi: 10.1049/iet-its.2014.0094.
[38] J. Li, J. Li, and H. Zhang, "Deep Learning Based Parking Prediction on Cloud Platform," in 2018 4th International Conference on Big Data Computing and Communications (BIGCOM), 7-9 Aug. 2018 2018, pp. 132-137, doi: 10.1109/BIGCOM.2018.00028.
[39] Y. Li, R. Yu, C. Shahabi, and Y. Liu, "Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting," presented at the ICLR 2018, 2018. [Online]. Available: https://doi.org/10.48550/arXiv.1707.01926.
[40] W. G. I. Limited. "A Brief History of Intelligent Transportation Systems." WJ Group Infrastructure Limited. https://www.wj.uk/advice/history-intelligent-transportation-systems/ (accessed May 9, 2025).
[41] T. Litman, "Parking Management: Comprehensive Implementation Guide," Victoria Transport Policy Institute, Victoria, Canada, January 22, 2025 2025. Accessed: 2025-05-08. [Online]. Available: http://www.vtpi.org/park_man_comp.pdf
[42] M. Liu, J. Naoum-Sawaya, Y. Gu, F. Lecue, and R. Shorten, "A Distributed Markovian Parking Assist System," IEEE Transactions on Intelligent Transportation Systems, vol. 20, no. 6, pp. 2230-2240, 2019, doi: 10.1109/TITS.2018.2865648.
[43] Q. Lu, Z. Tang, Y. Nie, and L. Peng, "ParkingRank-D: A Spatial-temporal Ranking Model of Urban Parking Lots in City-wide Parking Guidance System," in 2019 IEEE Intelligent Transportation Systems Conference (ITSC), 27-30 Oct. 2019 2019, pp. 388-393, doi: 10.1109/ITSC.2019.8917481.
[44] K. McCoy. "Parking pain causes financial and personal strain." https://www.usatoday.com/story/money/2017/07/12/parking-pain-causes-financial-and-personal-strain/467637001/ (accessed 7/4, 2025).
[45] L. E. a. R. M. a. P. S. a. K. W. a. M. Michalski, "SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference," presented at the ICLR 2020, 2020. [Online]. Available: https://arxiv.org/abs/1910.06591.
[46] Milesight. "AI Parking Management Pro Bullet Plus Camera." Milesight. https://www.milesight.com/security/product/ai-parking-management-pro-bullet-plus-camera (accessed.
[47] L. E. Y. K. Mimbela, L. A., "Summary of Vehicle Detection and Surveillance Technologies used in Intelligent Transportation Systems," United States. Department of Transportation. Federal Highway Administration, Washington, D.C., FHWA-HRT-08-103, 2007. [Online]. Available: https://rosap.ntl.bts.gov/view/dot/50558
[48] E. Natterer, R. Engelhardt, S. Hörl, and K. Bogenberger, "Graph Neural Network Approach to Predict the Effects of Road Capacity Reduction Policies: A Case Study for Paris, France," Computational Engineering, Finance, and Science (cs.CE), 2024, doi: arXiv:2408.06762.
[49] J. C. Provoost, A. Kamilaris, L. J. J. Wismans, S. J. van der Drift, and M. van Keulen, "Predicting parking occupancy via machine learning in the web of things," Internet of Things, vol. 12, p. 100301, 2020/12/01/ 2020, doi: https://doi.org/10.1016/j.iot.2020.100301.
[50] A. Sakai, K. Mizuno, T. Sugimoto, and T. Okuda, "Parking guidance and information systems," presented at the Pacific Rim TransTech Conference. 1995 Vehicle Navigation and Information Systems Conference Proceedings. 6th International VNIS. A Ride into the Future, 1995.
[51] W. Shao, Y. Zhang, B. Guo, K. Qin, J. Chan, and F. D. Salim, "Parking Availability Prediction with Long Short Term Memory Model," in Green, Pervasive, and Cloud Computing, Cham, S. Li, Ed., 2019// 2019: Springer International Publishing, pp. 124-137.
[52] W. Shao et al., "Transferrable contextual feature clusters for parking occupancy prediction," Pervasive and Mobile Computing, vol. 97, p. 101831, 2024/01/01/ 2024, doi: https://doi.org/10.1016/j.pmcj.2023.101831.
[53] D. C. Shoup, "Cruising for parking," Transport Policy, vol. 13, no. 6, pp. 479-486, 2006/11/01/ 2006, doi: https://doi.org/10.1016/j.tranpol.2006.05.005.
[54] D. C. Shoup, "The High Cost of Free Parking," Journal of Planning Education and Research, vol. 17, no. 1, pp. 3-20, 1997, doi: 10.1177/0739456x9701700102.
[55] A. Shvetsov, "Reducing the environmental impact of transport by reducing the time it takes to find parking for a car in a metropolis," presented at the E3S Web of Conferences, 2024. [Online]. Available: https://doi.org/10.1051/e3sconf/202451503009.
[56] M. T. J. Spaan, "Partially Observable Markov Decision Processes," in Reinforcement Learning: State-of-the-Art, M. Wiering and M. van Otterlo Eds. Berlin, Heidelberg: Springer Berlin Heidelberg, 2012, pp. 387-414.
[57] E. C. f. Transportation, "Webinar: Paved Paradise: How Parking Explains the World," in YouTube Webinar, ed. Eno Center for Transportation, 2023.
[58] E. I. Vlahogianni, K. Kepaptsoglou, V. Tsetsos, and M. G. Karlaftis, "A Real-Time Parking Prediction System for Smart Cities," Journal of Intelligent Transportation Systems, vol. 20, no. 2, pp. 192-204, 2016/01/01/ 2016, doi: https://doi.org/10.1080/15472450.2015.1037955.
[59] Y. Wang, Z. Chen, K. Zhang, Y. Cui, Y. Yang, and L. Peng, "Efficient Large-Scale Urban Parking Prediction: Graph Coarsening Based on Real-Time Parking Service Capability," IEEE Intelligent Transportation Systems Magazine, pp. 2-25, 2025, doi: 10.1109/MITS.2025.3532667.
[60] X. Xia and G. Gao, "Optimization model of parking space allocation based on genetic algorithm in reservation system mode," BCP Business & Management, vol. 29, pp. 21-27, %10/%12 2022, doi: 10.54691/bcpbm.v29i.2162.
[61] X. Xiao, Z. Jin, Y. Hui, Y. Xu, and W. Shao, "Hybrid Spatial–Temporal Graph Convolutional Networks for On-Street Parking Availability Prediction," Remote Sensing, vol. 13, no. 16, p. 3338, 2021. [Online]. Available: https://www.mdpi.com/2072-4292/13/16/3338.
[62] X. Xiao et al., "Parking Prediction in Smart Cities: A Survey," IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 10, pp. 10302-10326, 2023, doi: 10.1109/TITS.2023.3279024.
[63] B. Xu, O. Wolfson, J. Yang, L. Stenneth, P. S. Yu, and P. C. Nelson, "Real-Time Street Parking Availability Estimation," presented at the 2013 IEEE 14th International Conference on Mobile Data Management, 2013. [Online]. Available: https://ieeexplore.ieee.org/stampPDF/getPDF.jsp?tp=&arnumber=6569118&ref=.
[64] J. Yang, J. Portilla, and T. Riesgo, "Smart parking service based on Wireless Sensor Networks," in IECON 2012 - 38th Annual Conference on IEEE Industrial Electronics Society, 25-28 Oct. 2012 2012, pp. 6029-6034, doi: 10.1109/IECON.2012.6389096. [Online]. Available: https://doi.org/10.1109/IECON.2012.6389096
[65] S. Yang, W. Ma, X. Pi, and S. Qian, "A deep learning approach to real-time parking occupancy prediction in transportation networks incorporating multiple spatio-temporal data sources," Transportation Research Part C: Emerging Technologies, vol. 107, pp. 248-265, 2019/10/01/ 2019, doi: https://doi.org/10.1016/j.trc.2019.08.010.
[66] Z. Yanxu, S. Rajasegarar, and C. Leckie, "Parking availability prediction for sensor-enabled car parks in smart cities," in 2015 IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), 7-9 April 2015 2015, pp. 1-6, doi: 10.1109/ISSNIP.2015.7106902.
[67] K.-H. Yeh, "Parking Space Probability Prediction and Route Planning Affected by Spatio-Temporal Information Fluctuation," Master Academic Thesis, Department of Computer Science, National Chengchi University, 2023. [Online]. Available: https://hdl.handle.net/11296/zw476r
[68] B. Yu, H. Yin, and Z. Zhu, "Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting," presented at the Proceedings of the 27th International Joint Conference on Artificial Intelligence, Stockholm, Sweden, 2018.
[69] C. Zeng, C. Ma, K. Wang, and Z. Cui, "Parking Occupancy Prediction Method Based on Multi Factors and Stacked GRU-LSTM," IEEE Access, vol. 10, pp. 47361-47370, 2022, doi: 10.1109/ACCESS.2022.3171330.
[70] F. Zhang et al., "PewLSTM: Periodic LSTM with Weather-Aware Gating Mechanism for Parking Behavior Prediction," presented at the Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, 2020. [Online]. Available: https://doi.org/10.24963/ijcai.2020/610.
[71] W. Zhang, H. Liu, Y. Liu, J. Zhou, and H. Xiong, "Semi-Supervised Hierarchical Recurrent Graph Neural Network for City-Wide Parking Availability Prediction," arXiv preprint, vol. arXiv:1911.10516, 2019. [Online]. Available: https://doi.org/10.48550/arXiv.1911.10516.
[72] X. Zhang, C. Zhao, F. Liao, X. Li, and Y. Du, "Online parking assignment in an environment of partially connected vehicles: A multi-agent deep reinforcement learning approach," Transportation Research Part C: Emerging Technologies, vol. 138, p. 103624, 2022/05/01/ 2022, doi: https://doi.org/10.1016/j.trc.2022.103624.
[73] D. Zhao et al., "MePark: Using Meters as Sensors for Citywide On-Street Parking Availability Prediction," IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 7, pp. 7244-7257, 2022, doi: 10.1109/TITS.2021.3067675.
[74] L. Zhao et al., "T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction," IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 9, pp. 3848-3858, 2020, doi: 10.1109/TITS.2019.2935152.
[75] X. Zhao and M. Zhang, "Enhancing Predictive Models for On-Street Parking Occupancy: Integrating Adaptive GCN and GRU with Household Categories and POI Factors," Mathematics, vol. 12, no. 18, 2024, doi: 10.3390/math12182823.