Advances in Autonomous Traffic Management and Multi-Agent Systems

The field of autonomous traffic management is moving towards the development of more sophisticated multi-agent systems that can effectively manage intricate interactions between diverse traffic participants. Researchers are exploring the application of game theory, reinforcement learning, and decentralized control strategies to optimize traffic signal coordination, reduce congestion, and improve overall traffic efficiency. Notable developments include the use of signal attenuation to enable scalable decentralized multi-agent reinforcement learning and the proposal of novel frameworks for managing large-scale mixed traffic networks. These advancements have the potential to revolutionize urban mobility and reduce traffic-related challenges. Noteworthy papers include:

  • The paper on decision making in urban traffic using a game theoretic approach, which proposes a rule-based framework for autonomous vehicle decision-making and planning.
  • The paper on signal attenuation enabling scalable decentralized multi-agent reinforcement learning, which provides a useful model for future extensions to additional problems in wireless communications and radar networks.

Sources

Decision Making in Urban Traffic: A Game Theoretic Approach for Autonomous Vehicles Adhering to Traffic Rules

Signal attenuation enables scalable decentralized multi-agent reinforcement learning over networks

Origin-Destination Pattern Effects on Large-Scale Mixed Traffic Control via Multi-Agent Reinforcement Learning

Multi-agent Reinforcement Learning vs. Fixed-Time Control for Traffic Signal Optimization: A Simulation Study

Optimizing Resource Allocation for QoS and Stability in Dynamic VLC-NOMA Networks via MARL

Arrival Control in Quasi-Reversible Queueing Systems: Optimization and Reinforcement Learning

Built with on top of