UK’s Aston University scientists have developed a new AI system that could make long queues at traffic lights a thing of the past.
According to the scientists, their first-of-its-kind system reads live camera footage and adapts the lights to compensate, keeping the traffic flowing and reducing congestion.
It does this by using deep reinforcement learning, where a program understands when it is not doing well and tries a different course of action or continues to improve when it makes progress.
According to the scientists, the main form of traffic light automation used today at junctions depends on magnetic induction loops.
This manually-designed phase transition involves a wire that sits on the road and registers cars passing over it. The program then counts that and reacts to the available data.
Unlike the above mentioned method, Aston’s AI ‘sees’ through high traffic volume before the cars have gone through the lights and makes its decision then. This makes it more responsive and reactive than any other method utilized so far.
“The reason we have based this program on learned behaviors is so that it can understand situations it hasn’t explicitly experienced before,” notes Aston senior lecturer Dr George Vogiatzis
“We’ve tested this with a physical obstacle that is causing congestion, rather than traffic light phasing, and the system still did well.”
“As long as there is a causal link, the computer will ultimately figure out what that link is. It’s an intensely powerful system.”
The scientists have also built a photo-realistic traffic simulator called Traffic 3D to teach their program on how to handle different traffic and weather scenarios.
According to the scientists, when the system was tested on a real junction, it subsequently adapted to real traffic intersections despite being trained entirely on simulations.
This shows that by inculcating learned behaviors, the new system can be effective in many real-world settings.
Image and content: Staboslaw-Pixabay/Aston University
