Deep Reinforcement Learning for Adaptive Traffic Signal Control

TL;DR

This paper presents a deep reinforcement learning framework for adaptive traffic signal control that considers realistic traffic scenarios and sensors, demonstrating improved performance over traditional methods in simulation.

Contribution

It introduces a novel DRL-based traffic control framework with a new reward function tailored for realistic conditions, bridging the gap to real-world deployment.

Findings

Significantly improved traffic flow performance with the new reward function.

Effective adaptation to realistic traffic scenarios and sensor data.

Validation through simulation demonstrates practical potential.

Abstract

Many existing traffic signal controllers are either simple adaptive controllers based on sensors placed around traffic intersections, or optimized by traffic engineers on a fixed schedule. Optimizing traffic controllers is time consuming and usually require experienced traffic engineers. Recent research has demonstrated the potential of using deep reinforcement learning (DRL) in this context. However, most of the studies do not consider realistic settings that could seamlessly transition into deployment. In this paper, we propose a DRL-based adaptive traffic signal control framework that explicitly considers realistic traffic scenarios, sensors, and physical constraints. In this framework, we also propose a novel reward function that shows significantly improved traffic performance compared to the typical baseline pre-timed and fully-actuated traffic signals controllers. The framework is implemented and validated on a simulation platform emulating real-life traffic scenarios and sensor data streams.