Traffic Congestion Monitoring System

Abstract

Metropolitan cities are experiencing high volumes of rapid urbanization, creating excessive traffic queues, delays, and pollutants. Many traffic management systems utilize static traffic models that cannot dynamically correlate changes in vehicle traffic to enable optimal use of road networks. In this paper, we propose a Smart Adaptive Traffic Management System that uses artificial intelligence (AI) to monitor, predict, and control vehicles in real-time. It uses a combination of computer vision techniques and machine learning methods to detect a vehicle, classify the traffic density type, and adapt signal timing to the traffic density type. Overall, the proposed additive model using AI will help to enable efficient traffic flow, reduced waiting time at intersections, and less fuel consumption and emission. We show examples of experiments and simulations to help demonstrate a new configuration for traffic management systems to create sustainable and intelligent mobility for urban areas.