As urban traffic congestion continues to worsen in metropolitan cities across the globe, cars keep piling up in long queues, fuel is wasted, and rescuers get delayed. The fixed-cycle system of the old conventional traffic lights fails to adapt to realtime traffic conditions due to the presence of peak hours or when an ambulance has to pass through it. So, these old-fashioned systems still prove to be the most ineffective because they cannot dynamically respond in a time when flow becomes increasingly dynamic along with emergency situations. Our solution includes an optimization system of traffic; the cutting-edge technologies we utilize are: first, real-time vehicle detection technology through YOLO (You Only Look Once), and, second, intelligent control of signals with the help of Reinforcement Learning (RL). It analyzes video streams from traffic cameras and determines the number and type of vehicles approaching intersections, and the signal timings are adjusted in real time by the RL model to minimize wait times and give priority to emergency vehicles. Simulation tests had optimistic results, with a considerable reduction in vehicle idle time and improved flow of traffic. Emergency vehicles pass through the intersections with greater speed, and overall congestions were decreased. Our approach results in 30% improvement in average wait times as compared to the basic system, showing efficiency and accuracy in traffic management.
Traffic Optimization and Automated Emergency Vehicle Response System
06.03.2025
574874 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
BASE | 2019
|Automated RFID-Based Traffic Regulation and Vehicle Alert System for Emergency Conditions
Springer Verlag | 2021
|Automated RFID-Based Traffic Regulation and Vehicle Alert System for Emergency Conditions
British Library Conference Proceedings | 2021
|