Like a beating heart, cars flow through the streets at different rates depending on the time of day, day of the week and special events or holidays, sometimes causing inconvenient traffic jams. Researchers including Dr. Guni Sharon from the Texas A&M University Department of Computer Science and Engineering are developing new methods to optimize traffic controllers and reduce wait times at intersections.
Today, most traffic lights utilize signal controllers programmed with time settings that automatically adjust depending on traffic movement and time of day. Sharon and his team are integrating an intelligent algorithm that will allow signal controllers to adapt and respond to current traffic conditions in real time.
Their approach utilizes machine learning to optimize decision-making, and it reduced vehicle delay by 19.4% in trials. However, it took two days for the controller to understand what actions meaningfully helped mitigate traffic congestion from all directions. As a result, Sharon said the team’s future work will examine techniques to jumpstart the controller’s learning process.