Abstract
Intelligent transportation systems aim at a more efficient solution to the problems of traffic management, traffic analysis and prediction, route guidance, and provide more coordinated and “smarter” use of transport infrastructure in general. In this study, we consider a traffic signal control problem, which is a challenging problem in the transportation area. Traffic signal control aims to optimize traffic flows in road networks, decrease travel and waiting times, and increase the effectiveness of transport infrastructure usage. In this paper, we developed a novel hybrid reinforcement learning-based approach to solve the traffic signal control problem. In the first step of the proposed approach, we predict the number of vehicles that will cross an intersection during a specified time interval using a deep learning approach. This factor is used as one of a component of the system state space. In the second step, we proposed to use a double Q-learning approach to solve the traffic signal control problem using the observed system state space. An experimental study of the proposed algorithm was conducted using an open-source microscopic traffic simulation package SUMO. The effectiveness of the proposed approach is evaluated using both synthetic and real-world traffic scenarios. Experiments demonstrated that the proposed algorithm outperforms other baseline classical and state-of-the-art reinforcement learning-based algorithms in terms of the average waiting time and average travel time criteria.
Similar content being viewed by others
REFERENCES
Silva, B.N., Khan, M., and Han, K., Towards sustainable smart cities: A review of trends, architectures, components, and open challenges in smart cities, Sustainable Cities Soc., 2018, vol. 38, pp. 697–713. https://doi.org/10.1016/j.scs.2018.01.053
Agafonov, A.A. and Yumaganov, A.S., Bus arrival time prediction using recurrent neural network with LSTM architecture, Opt. Mem. Neural Networks, 2019, vol. 28, pp. 222–230. https://doi.org/10.3103/S1060992X19030081
Lim, C., Kim, K.J., and Maglio, P.P., Smart cities with big data: Reference models, challenges, and considerations, Cities, 2018, vol. 82, pp. 86–99. https://doi.org/10.1016/j.cities.2018.04.011
Kandt, J. and Batty, M., Smart cities, big data and urban policy: Towards urban analytics for the long run, Cities, 2021, vol. 109, p. 102992. https://doi.org/10.1016/j.cities.2020.102992
Ismagilova, E., Hughes, L., Dwivedi, Y.K., and Raman, K.R.: Smart cities: Advances in research – An information systems perspective, Int. J. Inf. Manage., 2019, vol. 47, pp. 88–100. https://doi.org/10.1016/j.ijinfomgt.2019.01.004
Schrank, D., Albert, L., Eisele, B., and Lomax, T., Urban Mobility Report, 2021.
Agafonov, A. and Yumaganov, A., Short-term traffic flow forecasting using a distributed spatial-temporal k nearest neighbors model, in Proceedings – 21st IEEE International Conference on Computational Science and Engineering (CSE 2018), 2018, pp. 91–98. https://doi.org/10.1109/CSE.2018.00019
Agafonov, A.A., Short-term traffic data forecasting: A deep learning approach, Opt. Mem. Neural Networks, 2021, vol. 30, no. 1, pp. 1–10. https://doi.org/10.3103/S1060992X21010021
Adart, A., Mouncif, H., and Na¨ımi, M., Vehicular ad-hoc network application for urban traffic management based on markov chains, Int. Arabic. J. Inf. Technol., 2017, vol. 14 (4A Spec. Issue), pp. 624–631.
Connected and Automated Vehicles: market forecast 2020. https://www.gov.uk/government/publications/connected-and-automated-vehicles-market-forecast-2020.
Allsop, R., Estimating the traffic capacity of a signalized road junction, Transp. Res., 1972, vol. 6, no. 3, pp. 245–255. https://doi.org/10.1016/0041-1647(72)90017-2
Webster, F.V., Traffic Signal Settings, H.M. Stationery Office, 195.)
Papageorgiou, M., Kiakaki, C., Dinopoulou, V., and Kotsialos, A., Yibing Wang: Review of road traffic control strategies, Proc. IEEE, 2003, vol. 91, no. 12, pp. 2043–2067. https://doi.org/10.1109/JPROC.2003.819610
Wei, H., Zheng, G., Gayah, V., and Li, Z., A Survey on Traffic Signal Control Methods. arXiv:1904.08117 [cs, stat] (2020). http://arxiv.org/abs/1904.08117, arXiv: 1904.08117.
Qadri, S.S.S.M., G¨ok¸ce, M.A., and Oner, E., State-of-art review of traffic signal control methods: challenges and opportunities, Eur. Transp. Res. Rev., 2020, vol. 12, no. 1, p. 55. https://doi.org/10.1186/s12544-020-00439-1
Guo, Q., Li, L., and (Jeff) Ban, X., Urban traffic signal control with connected and automated vehicles: A survey, Transp. Res., Part C: Emerging Technol., 2019, vol. 101, pp. 313–334. https://doi.org/10.1016/j.trc.2019.01.026
Little, J., Kelson, M., and Gartner, N., MAXBAND: A Program for Setting Signals on Arteries and Triangular Networks, Transp. Res. Rec. J. Transp. Res. Board, 1981, vol. 795, pp. 40–46.
Li, M.T. and Gan, A., Signal timing optimization for oversaturated networks using TRANSYT-7F, Transp. Res. Rec., 1999, vol. 1683, pp. 118–126. https://doi.org/10.3141/1683-15
Varaiya, P., The max-pressure controller for arbitrary networks of signalized intersections, in Advances in Dynamic Network Modeling in Complex Transportation Systems, Ukkusuri, S.V. and Ozbay, K., Eds., Complex Networks and Dynamic Systems, New York: Springer, 2013, pp. 27–66. https://doi.org/10.1007/978-1-4614-6243-92.
Yau, K.L., Qadir, J., Khoo, H., Ling, M., and Komisarczuk, P., A survey on Reinforcement learning models and algorithms for traffic signal control, ACM Comput. Surv., 2017, vol. 50, no. 3. https://doi.org/10.1145/3068287
Greguri’c, M., Vuji’c, M., Alexopoulos, C., and Mileti’c, M., Application of deep reinforcement learning in traffic signal control: An overview and impact of open traffic data, Appl. Sci., 2020, vol. 10, no. 11, p. 4011. https://doi.org/10.3390/app10114011
Palos, P. and Huszak, A., Comparison of Q-learning based traffic light control methods and objective functions, in 2020 International Conference on Software, Telecommunications and Computer Networks (SoftCOM), IEEE, Split, Hvar, Croatia, 2020, pp. 1–6. https://doi.org/10.23919/SoftCOM50211.2020.9238290.
Wei, H., Zheng, G., Yao, H., and Li, Z., IntelliLight: A reinforcement learning approach for intelligent traffic light control, in Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London United Kingdom: ACM, 2018, pp. 2496–2505. https://doi.org/10.1145/3219819.3220096.
Wei, H., Xu, N., Zhang, H., Zheng, G., Zang, X., Chen, C., Zhang, W., Zhu, Y., Xu, K., and Li, Z., CoLight: Learning network-level cooperation for traffic signal control, Proceedings of the 28th ACM International Conference on Information and Knowledge Management, 2019, pp. 1913–1922. https://doi.org/10.1145/3357384.3357902, http://arxiv.org/abs/1905.05717. arXiv: 1905.05717.
Chen, C., Wei, H., Xu, N., Zheng, G., Yang, M., Xiong, Y., Xu, K., and Li, Z., Toward a thousand lights: Decentralized deep reinforcement learning for large-scale traffic signal control, Proceedings of the AAAI Conference on Artificial Intelligence 34(04), 2020, pp. 3414–3421. https://doi.org/10.1609/aaai.v34i04.5744
Liu, Y., Liu, L., and Chen, W.P., Intelligent Traffic Light Control Using Distributed Multi-agent Q Learning. arXiv:1711.10941[cs], 2017. http://arxiv.org/abs/1711.10941. arXiv: 1711.10941.
Li, Z., Yu, H., Zhang, G., Dong, S., and Xu, C.Z., Network-wide traffic signal control optimization using a multi-agent deep reinforcement learning, Transp. Res., Part C: Emerging Technol., 2021, vol. 125, p. 103059. https://doi.org/10.1016/j.trc.2021.103059
Gu, J., Fang, Y., Sheng, Z., and Wen, P., Double deep Q-network with a DualAgent for traffic signal control, Appl. Sci., 2020, vol. 10, no. 5, p. 1622. https://doi.org/10.3390/app10051622
Agafonov, A. and Myasnikov, V., Traffic Signal Control: A Double Q-learning Approach, in Proceedings of the 16th Conference on Computer Science and Intelligence Systems, FedCSIS 2021, 2021, pp. 365–369. https://doi.org/10.15439/2021F109.
Zeng, J., Hu, J., and Zhang, Y., Adaptive traffic signal control with deep recurrent Q-learning, in 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu: IEEE, 2018, pp. 1215–1220. https://doi.org/10.1109/IVS.2018.8500414.
Lopez, P.A., Wiessner, E., Behrisch, M., Bieker-Walz, L., Erdmann, J., Flotterod, Y.P., Hilbrich, R., Lucken, L., Rummel, J., and Wagner, P., Microscopic traffic simulation using SUMO, in 2018 21st International Conference on Intelligent Transportation Systems (ITSC), Maui, HI: IEEE, 2018, pp. 2575–2582. https://doi.org/10.1109/ITSC.2018.8569938.
Hasselt, H., Double Q-learning, Adv. Neural Inform. Process. Syst., 2010, vol. 23.
RESCO, 2021. https://github.com/Pi-Star-Lab/RESCO, original-date: 2021-06-07T17:31:48Z.
TAPASCologne-SUMO Documentation, https://sumo.dlr.de/docs/Data/Scenarios/TAPASCologne.html.
Ault, J., Hanna, J.P., and Sharon, G., Learning an Interpretable Traffic Signal Control Policy. arXiv:1912.11023 [cs, stat], 2020. http://arxiv.org/abs/1912.11023. arXiv: 1912.11023.
Funding
The work was supported by the Russian Science Foundation, grant no. 21-11-00321, https://rscf.ru/en/project/21-11-00321/.
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
The authors declare that they have no conflicts of interest.
About this article
Cite this article
Agafonov, A.A., Myasnikov, V.V. Hybrid Prediction-Based Approach for Traffic Signal Control Problem. Opt. Mem. Neural Networks 31, 277–287 (2022). https://doi.org/10.3103/S1060992X2203002X
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1060992X2203002X