Smart Signal Control Using Reinforcement Learning to Ease Urban Traffic

Sivakumar Ponnusamy, M. L. M. Prasad, G. Swarnalakshmi, S. Karthikeyan, S. K. Lokesh Naik, Moses Raja A.

2025

Abstract

This study offers an innovative proposal for the management urban traffic through the implementation of reinforcement learning for adaptive traffic signal systems to overcome the traffic jam problem and increase vehicle flow. The model, based on deep learning algorithms and real-time data, dynamically modifies timing in accordance with traffic conditions, facilitating smooth and hang-up-free traffic. Compared to classical static or rule-based methodologies, the approach is flexible enough to adapt to changing patterns and is shown that outperforms in simulations based on realistic urban networks. In addition, the system uses multi-agent cooperation methodology to optimize signal coordination across intersections, thus being scalable and responsive. The results show that the learning instantiation can lead to significant gains in traffic efficiency, delay reduction, and emission mitigation, demonstrating the potential of reinforcement learning as a promising approach to enhance intelligent transportation systems.

Paper Citation

in Harvard Style

Ponnusamy S., Prasad M., Swarnalakshmi G., Karthikeyan S., Naik S. and A. M. (2025). Smart Signal Control Using Reinforcement Learning to Ease Urban Traffic. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 102-108. DOI: 10.5220/0013858100004919

in Bibtex Style

@conference{icrdicct`2525,
author={Sivakumar Ponnusamy and M. Prasad and G. Swarnalakshmi and S. Karthikeyan and S. Naik and Moses A.},
title={Smart Signal Control Using Reinforcement Learning to Ease Urban Traffic},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25},
year={2025},
pages={102-108},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013858100004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - Volume 1: ICRDICCT`25
TI - Smart Signal Control Using Reinforcement Learning to Ease Urban Traffic
SN - 978-989-758-777-1
AU - Ponnusamy S.
AU - Prasad M.
AU - Swarnalakshmi G.
AU - Karthikeyan S.
AU - Naik S.
AU - A. M.
PY - 2025
SP - 102
EP - 108
DO - 10.5220/0013858100004919
PB - SciTePress