Intelligent traffic light system with real-time visual analysis and clean energy with priority for emergency vehicles

Abstract

In Panama’s urban areas, traffic management presents a recurring problem: traffic lights operate with fixed cycles, without taking into account the real-time flow of vehicles or emergency situations. This lack of adaptability causes unnecessary congestion, wasted time, increased fuel consumption and higher pollutant emissions. The inefficiency of the current system reflects the urgent need for technological solutions to optimize urban mobility. In response to this challenge, the present project proposes the development of a functional prototype of an intelligent and self-sustainable traffic light, designed to dynamically adapt to traffic conditions. The system employs a camera connected to an artificial intelligence model capable of analyzing images in real time. Through the YOLO (You Only Look Once) model, it detects the number of vehicles on each lane and automatically adjusts the traffic light cycles to optimize the flow. In addition, microphones are integrated to identify siren sounds, giving priority passage to ambulances and emergency vehicles. This approach not only improves road efficiency but also enhances public safety. The physical prototype was built with an Arduino board and LED lights to simulate signaling, while the processing was developed in Python using YOLO along with cameras for traffic detection and developing logic for emergency priority, and is solar powered, reinforcing its commitment to environmental sustainability.