Smart Traffic Management System Using IoT and Machine Learning for Real-Time Adaptive Signal Control

Author(s): Vaseekurrahman, Neeraj Kumar and Noorishta Hashmi

Abstract:

Rapid urbanization and exponential growth in vehicular population have intensified traffic congestion in metropolitan and developing cities. Traditional traffic signal systems operate on fixed time intervals without considering real-time vehicle density, leading to inefficient traffic flow, fuel wastage, and increased carbon emissions. This paper presents a Smart Traffic Management System integrating Internet of Things (IoT), computer vision, and machine learning for adaptive signal control. The system utilizes YOLO-based vehicle detection to estimate real-time lane density and dynamically adjust green signal duration. Experimental evaluation demonstrates reduced waiting time and improved traffic throughput compared to conventional systems, supporting scalable smart city deployment. Additionally, the architecture supports real-time data aggregation and predictive analytics to enhance signal optimization. Simulation results confirm improved responsiveness, scalability, and reliability under fluctuating traffic volumes in urban environments. The framework also enables future integration with autonomous vehicle communication systems.

keywords:

Pages: 122-125     |    12 View     |    1 Download