Researchers at Aston University in Birmingham, England, have designed a new artificial intelligence system that could put an end to long queues at traffic lights, according to a statement.
The The system reads live camera feeds and adapts traffic lights accordingly to keep traffic moving in the best possible way and reduce congestion. The system – the first of its kind, according to scientists – is so effective because it uses deep reinforcement learning, which means it adapts its processes when it doesn’t get it right, tries a different course of action, and keeps getting better as it progresses.
Photorealistic traffic simulator
In testing, the system far outperformed all other methods, which typically rely on manually designed phase transitions. This is due, they explain, to the fact that was built on top of a state-of-the-art photorealistic traffic simulator called Traffic 3D.
The developers, who published the results of their study in the journal Aston University’s Library ServicesThey hope to begin real road testing later this year.
“We have set it up as a traffic control game. The program receives a ‘reward’ when it gets a car through a junction,” explained Dr Maria Chli, Reader in Computer Science at Aston University.
“Any time a car has to wait or there is a traffic jam, there is a negative reward. Actually, there is no input from us; we just control the reward system,” she added.
Adaptable to different traffic and weather scenarios
The simulator has also been trained to handle different traffic and weather scenarios and was therefore able to quickly adapt to real traffic intersections, making it effective in many real-world scenarios.
“The reason we have based this program on learned behaviors is so that you can understand situations that you have not explicitly experienced before. We have tested it with a physical obstacle causing the congestion, instead of the traffic light phases, and the system continued to work fine. As long as there is a causal link, the computer will eventually figure out what that link is. It’s an intensely powerful system,” concluded Dr George Vogiatzis, Senior Lecturer in Computer Science at Aston University. (I)