Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Smart Traffic Signals: Comparing MARL and Fixed-Time Strategies

Saahil Mahato

TL;DR

This work investigates whether multi-agent reinforcement learning can outperform traditional fixed-time traffic signals in a multi-intersection network. Using a Pygame-based simulator, it deploys decentralized DQN agents at each signal and compares them against a fixed-time baseline across multiple randomized runs. The results show MARL achieves statistically significant improvements in throughput and, more notably, dramatic reductions in average wait times, indicating strong potential for adaptive urban traffic management. The study also discusses limitations and outlines paths toward real-world validation and scalability enhancements to inform deployment in smart cities.

Abstract

Urban traffic congestion, particularly at intersections, significantly affects travel time, fuel consumption, and emissions. Traditional fixed-time signal control systems often lack the adaptability to effectively manage dynamic traffic patterns. This study explores the application of multi-agent reinforcement learning (MARL) to optimize traffic signal coordination across multiple intersections within a simulated environment. A simulation was developed to model a network of interconnected intersections with randomly generated vehicle flows to reflect realistic traffic variability. A decentralized MARL controller was implemented in which each traffic signal operates as an autonomous agent, making decisions based on local observations and information from neighboring agents. Performance was evaluated against a baseline fixed-time controller using metrics such as average vehicle wait time and overall throughput. The MARL approach demonstrated statistically significant improvements, including reduced average waiting times and improved throughput. These findings suggest that MARL-based dynamic control strategies hold substantial promise to improve urban traffic management efficiency. More research is recommended to address the challenges of scalability and real-world implementation.

Smart Traffic Signals: Comparing MARL and Fixed-Time Strategies

TL;DR

This work investigates whether multi-agent reinforcement learning can outperform traditional fixed-time traffic signals in a multi-intersection network. Using a Pygame-based simulator, it deploys decentralized DQN agents at each signal and compares them against a fixed-time baseline across multiple randomized runs. The results show MARL achieves statistically significant improvements in throughput and, more notably, dramatic reductions in average wait times, indicating strong potential for adaptive urban traffic management. The study also discusses limitations and outlines paths toward real-world validation and scalability enhancements to inform deployment in smart cities.

Abstract

Urban traffic congestion, particularly at intersections, significantly affects travel time, fuel consumption, and emissions. Traditional fixed-time signal control systems often lack the adaptability to effectively manage dynamic traffic patterns. This study explores the application of multi-agent reinforcement learning (MARL) to optimize traffic signal coordination across multiple intersections within a simulated environment. A simulation was developed to model a network of interconnected intersections with randomly generated vehicle flows to reflect realistic traffic variability. A decentralized MARL controller was implemented in which each traffic signal operates as an autonomous agent, making decisions based on local observations and information from neighboring agents. Performance was evaluated against a baseline fixed-time controller using metrics such as average vehicle wait time and overall throughput. The MARL approach demonstrated statistically significant improvements, including reduced average waiting times and improved throughput. These findings suggest that MARL-based dynamic control strategies hold substantial promise to improve urban traffic management efficiency. More research is recommended to address the challenges of scalability and real-world implementation.

Paper Structure

This paper contains 56 sections, 4 equations, 4 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Research Objective
  3. Related Work
  4. Fixed-Time and Actuated Signal Control
  5. Adaptive Traffic Control Systems (ATCS)
  6. Artificial Intelligence in Traffic Signal Control
  7. Reinforcement Learning Approaches
  8. Multimodal and Safety-Oriented Traffic Control
  9. Future Directions
  10. Summary
  11. Methodology
  12. Simulation Environment
  13. Simulation Framework
  14. Road Network Configuration
  15. Traffic Signal Parameters
  16. ...and 41 more sections

Figures (4)

  • Figure 1: Screenshot of the Pygame-based traffic simulation environment
  • Figure 2: Deep Q-Network Architecture
  • Figure 3: State Representation
  • Figure 4: Full Training Procedure