Learning Decentralized Traffic Signal Controllers With Multi-Agent Graph Reinforcement Learning
IEEE Transactions on Mobile Computing · 2024-06 · 23 (6) · 7180-7195
DOI: 10.1109/tmc.2023.3332081 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2024-06. Verfasst von Yao Zhang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 33-mal zitiert (Crossref) und nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Science Fund for Distinguished Young Scholars, Hong Kong Research Grants Council.
Abstract (Originalsprache)
This paper considers optimal traffic signal control in smart cities, which has been taken as a complex networked system control problem. Given the interacting dynamics among traffic lights and road networks, attaining controller adaptivity and scalability stands out as a primary challenge. Capturing the spatial-temporal correlation among traffic lights under the framework of Multi-Agent Reinforcement Learning (MARL) is a promising solution. Nevertheless, existing MARL algorithms ignore effective information aggregation which is fundamental for improving the learning capacity of decentralized agents. In this paper, we design a new decentralized control architecture with improved environmental observability to capture the spatial-temporal correlation. Specifically, we first develop a topology-aware information aggregation strategy to extract correlation-related information from unstructured data gathered in the road network. Particularly, we transfer the road network topology into a graph shift operator by forming a diffusion process on the topology, which subsequently facilitates the construction of graph signals. A diffusion convolution module is developed, forming a new MARL algorithm, which endows agents with the capabilities of graph learning. Extensive experiments based on both synthetic and real-world datasets verify that our proposal outperforms existing decentralized algorithms.
Förderung
- National Natural Science Foundation of China
- National Science Fund for Distinguished Young Scholars
- Hong Kong Research Grants Council
- Agency for Science, Technology and Research
- RIE2020 Advanced Manufacturing and Engineering
Zitieren
APA
Zhang, Y., Yu, Z., Zhang, J., Wang, L., Luan, T. H., Guo, B., & Yuen, C. (2024). Learning Decentralized Traffic Signal Controllers With Multi-Agent Graph Reinforcement Learning. IEEE Transactions on Mobile Computing, 23(6), 7180-7195. https://doi.org/10.1109/tmc.2023.3332081
Vancouver
Zhang Y, Yu Z, Zhang J, Wang L, Luan TH, Guo B, et al. Learning Decentralized Traffic Signal Controllers With Multi-Agent Graph Reinforcement Learning. IEEE Transactions on Mobile Computing. 2024;23(6):7180-7195. doi:10.1109/tmc.2023.3332081
BibTeX
@article{Zhang2024,
author = {Zhang, Yao and Yu, Zhiwen and Zhang, Jun and Wang, Liang and Luan, Tom H. and Guo, Bin and Yuen, Chau},
title = {Learning Decentralized Traffic Signal Controllers With Multi-Agent Graph Reinforcement Learning},
journal = {IEEE Transactions on Mobile Computing},
year = {2024},
volume = {23},
number = {6},
pages = {7180--7195},
doi = {10.1109/tmc.2023.3332081}
}
EUVIMED