Fairness-aware Multi-agent Reinforcement Learning and Visual Perception For Adaptive Traffic Signal Control
DOI:
Author:
Affiliation:

Dalian Maritime University

Clc Number:

Fund Project:

no fund

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    The majority of Multi-agent Reinforcement Learning (MARL) methods for solving Adaptive Traffic Signal Control (ATSC) problems are dedicated to maximizing the throughput while ignoring fairness, resulting in a bad situation where some vehicles keep waiting. For this reason, this paper models the ATSC problem as a Partially Observable Markov Game (POMG), in which a value function that combines throughput and fairness is elaborated. On this basis, we propose a new cooperative MARL method FA-MAPPO, i.e., fairness-aware multi-agent proximity policy optimization, which is based on the cooperative MARL algorithm MAPPO. In addition, FA-MAPPO uses graph attention neural networks to efficiently extract state representations from traffic data acquired through visual perception in multi-intersection scenarios. Experimental results in Jinan and synthetic scenarios confirm that FA-MAPPO improves fairness while guaranteeing passage efficiency compared to the SOTA methods.

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:November 28,2023
  • Revised:January 15,2024
  • Adopted:January 29,2024
  • Online:
  • Published: