A Comparative Study of Algorithms for Intelligent Traffic Signal Control

TL;DR

This study compares various algorithms, including reinforcement learning methods, for optimizing traffic signal control to reduce waiting times and improve traffic flow using simulation models.

Contribution

It introduces a comprehensive comparison of traditional and reinforcement learning algorithms for traffic signal control within a simulated environment and real-world scenario.

Findings

Reinforcement learning methods outperform traditional scheduling.

Deep Q Network and A2C show significant improvements in traffic flow.

Simulation results validate effectiveness in real-world intersection.

Abstract

In this paper, methods have been explored to effectively optimise traffic signal control to minimise waiting times and queue lengths, thereby increasing traffic flow. The traffic intersection was first defined as a Markov Decision Process, and a state representation, actions and rewards were chosen. Simulation of Urban MObility (SUMO) was used to simulate an intersection and then compare a Round Robin Scheduler, a Feedback Control mechanism and two Reinforcement Learning techniques - Deep Q Network (DQN) and Advantage Actor-Critic (A2C), as the policy for the traffic signal in the simulation under different scenarios. Finally, the methods were tested on a simulation of a real-world intersection in Bengaluru, India.