acm-header
Sign In

Communications of the ACM

ACM Careers

AI Traffic Light System Optimizes Flow in Complex Environments


traffic flowing during snowfall, illustration

The agent learns to optimize traffic flows in the presence of snow.

Credit: Aston University

Long queues at traffic lights could be a thing of the past, thanks to an artificial intelligence system developed by Aston University researchers.

The system reads live camera footage and adapts traffic lights to compensate, keeping vehicles flowing and reducing congestion. The researchers described their work in a paper presented at AAMAS 2022, the International Conference on Autonomous Agents and Multi-Agent Systems.

The team developed a fully-autonomous, vision-based deep reinforcement learning agent that achieves adaptive signal control in the face of complex, imprecise, and dynamic traffic environments. In testing, the system significantly outperformed all other methods. 

The researchers built a traffic simulator and set up a traffic control game. "The program gets a 'reward' when it gets a car through a junction," says Maria Chli of Aston's Computer Science Research Group. "Every time a car has to wait or there's a jam, there's a negative reward."

From Aston University
View Full Article


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account