Multi-Agent Deep Reinforcement Learning for Decentralized Cooperative Traffic Signal Control

Abstract

The traffic congestion becomes a severe problem in almost every city, and intelligent transportation systems make it possible for an adaptive traffic signal control system to improve signal control. Exploiting deep reinforcement learning for traffic signal control is a frontier topic in intelligent transportation research. However, it's hard to use centralized reinforcement learning for large-scale traffic signal control systems due to the high dimensions of the joint action space. Multi-agent deep reinforcement learning overcomes the curse of dimensions but introduces a new problem: how to learn coordination between agents under a partially observable traffic environment. In this paper, we introduce a multi-agent deep reinforcement learning algorithm for a large-scale traffic signal control system. The proposed method is compared with greedy policy, independent Q-learning method, and independent actor critic method in a large synthetic traffic networks. The simulation demonstrates the proposed method is more efficient than other decentralized reinforcement learning approaches.