ADAPTIVE AI BASED TRAFFIC MANAGEMENT SYSTEM

This research demonstrates an Adaptive AI-Based Traffic Management System that creates more efficient traffic
flow throughout cities while providing emergency vehicles with first priority. There is a simulated four-lane
intersection with Traffic Lights that are continuously monitored via video streaming and an associated still
image. A series of pictures as each car passes through the intersection will be taken with a near simultaneous
still image that serves as a frame for YOLO object detection. The TM system will analyze each lane to determine
the density of traffic utilizing the lane, and then it will adjust the timing of all traffic signals based on the density
of traffic in each lane. Additionally, the TM system will utilize optimization of each lane to provide ambulance
services access to intersections before any other vehicles. If the density of vehicles exceeds a predetermined
number, the TM system will start a timer to keep traffic moving efficiently. The TM system is expected to
decrease fuel consumption and vehicle emissions by reducing the length of time that vehicles wait at red lights,
thus contributing to the environment.