Dynamic AI Traffic Signal System for Real-Time Traffic Management Using Pygame and YOLO V8

Ruppa Ranjit Raj, Riya Shanker, Maranco M.

2025

Abstract

Urban traffic congestion leads to longer travel time, fuel consumption, and air pollution. Traditional traffic lights have fixed timing that cannot be tailored to the actual traffic condition, causing inefficiency and delay. This paper presents an AI-based Smart Traffic Management System (STMS) with optimized traffic flow through computer vision-based vehicle detection and an AI-based decision system to dynamically adjust signals. The network, by learning in real-time traffic congestion patterns, eliminates congestion points, shortens waiting time, and enhances urban mobility's combined traffic cameras, IoT sensors, and real-time analysis of data combines to estimate traffic density. Using deep learning for detection and reinforcement learning to fine-tune the signals, it optimizes traffic movement. Cost-efficient and scalable in relation to fixed installations, it adapts to urban infrastructure, lowering delays, fuel consumption, and emissions. This work introduces the shortcomings of conventional systems, summarizes intelligent traffic management studies, and discusses STMS structure and influence, suggesting a possible AI-based solution for modern cities.

Paper Citation

in Harvard Style

Raj R., Shanker R. and M. M. (2025). Dynamic AI Traffic Signal System for Real-Time Traffic Management Using Pygame and YOLO V8. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 300-310. DOI: 10.5220/0013912100004919

in Bibtex Style

@conference{icrdicct`2525,
author={Ruppa Raj and Riya Shanker and Maranco M.},
title={Dynamic AI Traffic Signal System for Real-Time Traffic Management Using Pygame and YOLO V8},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={300-310},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013912100004919},
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 - ICRDICCT`25
TI - Dynamic AI Traffic Signal System for Real-Time Traffic Management Using Pygame and YOLO V8
SN - 978-989-758-777-1
AU - Raj R.
AU - Shanker R.
AU - M. M.
PY - 2025
SP - 300
EP - 310
DO - 10.5220/0013912100004919
PB - SciTePress