| Abstrak/Abstract |
Traffic congestion at intersections is a major challenge in urban mobility that leads to increased travel time, fuel consumption, and exhaust emissions. Traditional traffic light control methods that rely on fixed time or simple sensors often cannot adapt to real-time traffic conditions. This research proposes an intelligent system for vehicle queue length detection and traffic light duration optimization by utilizing computer vision and artificial intelligence. The You Only Look Once (YOLO) algorithm is applied to detect vehicles as well as measure queue length from CCTV footage, thus enabling real-time traffic data capture. The obtained traffic parameters are then optimized using the Artificial Bee Colony (ABC) algorithm to dynamically adjust the green light duration, with the aim of minimizing queue length and waiting time and improving overall traffic efficiency. The effectiveness of the system is tested through simulation by comparing the fixed time-based control method with the proposed adaptive approach. The expected outcome of this research is improved traffic flow management and reduced congestion at intersections, thus contributing to a more sustainable urban transportation system. |
