Scalable multi-region perimeter metering control for urban networks: A multi-agent deep reinforcement learning approach

• Published in: Transportation research. Part C, Emerging technologies Vol. 148; p. 104033
• Main Authors: Zhou, Dongqin, Gayah, Vikash V.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2023
• Subjects: Macroscopic Fundamental Diagram (MFD); Model-free multi-agent reinforcement learning (MARL); Multi-region perimeter metering control; Multi-region perimeter metering control; Model-free multi-agent reinforcement learning (MARL); Macroscopic Fundamental Diagram (MFD)
• ISSN: 0968-090X; 1879-2359
• DOI: 10.1016/j.trc.2023.104033

•A model-free control scheme proposed for multi-region perimeter metering control•Scalability demonstrated via numerical simulations on a seven-region urban network•Learning and control efficay illustrated via comparisons to the MPC method•Resilience and transferability shown considering environment...