Adaptive Cooperative Control of Multi-traffic Signal Based on Multi-agent Reinforcement Learning

XVII International Conference on Systems, Automatic Control and Measurements, SAUM 2024 (pp. 1-5)

АУТОР(И) / AUTHOR(S): Jianxun Cui , Boyuan Zhao 

Download Full Pdf  

DOI:  10.46793/SAUM24.001C

САЖЕТАК / ABSTRACT:

Horizontal traffic signal intersection are the most common bottleneck areas in urban traffic operation. Due to safety conflicts between traffic from different directions at the same time and space node, only certain directions of traffic can be released within a limited time, while other directions of traffic are forced to stop and wait for red lights. Therefore, unreasonable signal timing can easily lead to significant delays and energy consumption in overall traffic operation. Although previous studies have proposed traffic signal and CAV control models based on optimization control, model predictive control, and reinforcement learning methods, they have difficulties to model the complex and dynamic scenarios for multi-intersection traffic network. Therefore, this paper proposes an adaptive and cooperative control model based on Multi-agent Reinforcement Learning (MARL) for traffic signal. A novel MARL algorithm and several simulation scenarios are developed to investigate traffic performance of the algorithm. The results show that the proposed cooperative model is effective, and it can significantly reduce the average queue length and improve traffic efficiency.

КЉУЧНЕ РЕЧИ / KEYWORDS:

multi-agent reinforcement learning, multi traffic signal intersection, decision-making and control

ПРОЈЕКАТ/ ACKNOWLEDGEMENT:

This research was supported by the joint guidance project of Heilongjiang Provincial Natural Science Foundation through Grant #LH2021E074 and the Fun-damental Research Funds for the Central Universities through Grant # HIT.NSRIF202235.

ЛИТЕРАТУРА / REFERENCES

  1. S. B. Cools, C. Gershenson and B. D’Hooghe, „Self-organizing traffic lights: A realistic simulation“, Springer London, 2006. DOI:10.1007/978-1-84628-982-8_3.
  2. Varaiya, Pravin, „Max pressure control of a network of signalized intersections“, IFAC-PapersOnLinenov.(2013):19-24.
  3. B. Hunt, D. I. Robertson, R. D. Bretherton, and M. C. Royle, „The SCOOT on-line traffic signal optimisation technique“, Traffic Eng., Control, vol. 23, no. 4, pp. 190–192, Apr. 1982.
  4. A. G. Sims, K. W. Dobinson, „SCAT the Sydney coordinated adaptive traffic system: philosophy and benefits“, International Symposium on Traffic Control Systems. Institution of Engineers, Australia, 1979.
  5. X. Cheng, Z.Yang, „Intelligent Traffic Signal Control Approach Based on Fuzzy-Genetic Algorithm“, Fuzzy Systems and Knowledge Discovery, 2008. FSKD ’08. Fifth International Conference on. IEEE, 2008.DOI:10.1109/FSKD.2008.389.
  6. D. Jia, Z. Chen, „Traffic Signal control optimization based on fuzzy neural network“, IEEE, 2012.DOI:10.1109/MIC.2012.6273473.