Abstract :

Rapid Urbanization And Increasing Vehicle Density Demand Intelligent Traffic Control Systems Capable Of Real-time Adaptation. Conventional Traffic Lights Operate On Fixed Timing Schedules That Fail To Reflect Dynamic Road Conditions, Resulting In Excessive Waiting Time, Fuel Wastage, And Increased Emissions. To Overcome These Limitations, This Project Proposes An Edge Machine Learning–driven Dynamic Traffic Light System (DTLS) That Performs Realtime Vehicle Detection And Adaptive Signal Control At The Intersection Level. The System Deploys An Optimized YOLObased Object Detection Model On Edge Devices To Identify And Count Vehicles From Live Video Streams. By Processing Data Locally, The Proposed Mechanism Significantly Reduces Latency And Network Dependency Compared To Centralized Architectures. A Hybrid Scheduling Strategy Integrating Shortest-job-first And Round-robin Principles Dynamically Allocates Green Signal Duration Based On Traffic Density Distribution. Additionally, Inter-junction Communication Using Low-power Wireless Protocols Enables Cumulative Delay Computation And Coordinated Traffic Prioritization Across Connected Intersections. A Smart Green Corridor Mechanism Is Incorporated To Ensure Uninterrupted Passage For Emergency Vehicles. The Proposed Architecture Enhances Scalability, Energy Efficiency, And Responsiveness, Making It Suitable For Smart City Deployment. Experimental Evaluation Demonstrates Improved Traffic Flow Management, Reduced Congestion, And Optimized Signal Utilization Under Varying Traffic Conditions. Keywords—Edge Computing, Intelligent Transportation Systems (ITS), Dynamic Traffic Light System (DTLS), Machine Learning, Object Detection, YOLO, Internet Of Things (IoT), IEEE 802.15.4, LoRaWAN, Low-power Wireless Networks, Traffic Congestion Control, Smart Cities, Vehicle Detection, Real-time Systems, Embedded Systems.

Published:

20-2-2026

Issue:

Vol. 26 No. 2 (2026)

Page Nos:

64-69

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite