Social Knowledge and Technology Combine to Improve Air Quality with Better Traffic Signals

A professor at Lehigh University's P.C. Rossin College of Engineering & Applied Science has been awarded a grant to work on improving air quality — by applying machine learning technology to traffic lights.

"This is the first project of its kind to incorporate a social component into a traffic control system," says Yu Yang, assistant professor of computer science and engineering, of the work for which the grant has been supplied. "We're taking both a technical and a social perspective to solve a real-world problem."

Yang's idea is to use more than just data on the flow of vehicles when deciding how traffic signals should operate. To start: data-gathering, using a low-cost mobile air-quality sensing system to pinpoint areas of high air pollution. These are then investigated for social requirements - such as the presence of a hospital, where it could be assumed that there will be more sensitive individuals than other areas.

"We'll use those data to then develop a spatial-temporal graph diffusion learning model to determine the traffic situation in our test-bed city of Newark, New Jersey," Yang explains. "In other words, what is both the traffic and the air pollution like at different points of time in different locations?"

A reinforcement learning system will then add in data from traffic signals around the city before running simulations to investigate how changes could improve air quality — providing evidence for the potential deployment of a traffic management system which would provide city managers real-time control while providing citizens with information on air quality levels in specific locations.

The idea ties in to another project in Yang's lab, investigating micromobility systems like electric bikes and scooters — attempting to make management of large-scale fleets, like those offered for hire on an hourly or daily basis, more efficient. "Cities can have thousands of these vehicles, and so the problem becomes managing them," he explains, "and making sure the location of the vehicles satisfies the demand for them."

The application of technology to real-world problems is at the heart of our upcoming Impact Summit, which takes place next week: a two-day developer summit which investigates real-world issues including the sustainable development of societies, how smart cities can improve citizens' lives, and how devices like the Nordic Thingy:53, Raven IoT Sensor Box, and the Wilderness Labs Meadow can be used to solve real-world problems.

Yang's research is still in progress, and has not yet been published; the Impact Summit, meanwhile, is now open for registrations.