Smart traffic lights save time and money

It is estimated that traffic congestion in urban areas in the UK caused residents to wait around 115 hours in 2019.

Of course, along with that is fuel waste and CO . emissions₂ pollute the environment.

“The main cause of congestion is the timing of the light signal is not appropriate, not smart enough” – specialized page Phys.org said.

In that context, researchers at Aston University (UK) built an AI-powered traffic model called Traffic 3D to teach their program to handle a variety of traffic and weather situations.

Simulate traffic light control by AI on the highway. Photo: Aston University

As a result, when tested on a real intersection, the system adapts itself to the intersections even though the training is based entirely on simulation. As a result, it can be effective in a real-life context.

Specifically, the AI ​​system can read images from real-time cameras and adjust lights to keep traffic flowing and reduce congestion. The system can know when it’s doing well or not so it can try something different or continue to improve accordingly.

In testing, the system fared better than any other method, which relies heavily on the operator to change the direction of the traffic lights.

“We’ve set up the simulation as a traffic control game. The program gets a ‘reward’ for getting the car through an intersection. Every time a car has to wait or theres a traffic jam, the program will deduct the reward. ” – said expert Maria Chli of the research team.

Currently, the form of automatic traffic lights used at intersections depends on the magnetic induction circuit to recognize passing vehicles. Meanwhile, the team’s AI technology “detects” high traffic before the vehicle passes through the lights and makes a decision, so it can react more quickly.