Sanjay Ranka, Ph.D., a professor in the Department of Computer & Information Science & Engineering (CISE), was recently awarded a $2 million grant from the National Science Foundation to develop technology that will monitor high-risk intersections in Gainesville, FL, to make roadways safer. The project, “Video Based Machine Learning for Smart Traffic Analysis and Management,” is a collaboration with the City of Gainesville and the UF Transportation Institute and will last four years.
The goal of the project is to use video processing and machine learning to make traffic intersections safer and improve congestion for the city. Ranka will be working with Anand Rangarajan, Ph.D., professor, CISE; Lily Elefteriadou, Ph.D., Director of the UF Transportation Institute (UFTI); Siva Srinivasan, Ph.D., associate professor, Engineering School of Sustainable Infrastructure & Environment; and Dan Hoffman, Gainesville Assistant City Manager.
“Gainesville, like other cities in Florida, has a number of pedestrian-related accidents,” Ranka said. “Many of them result in fatalities. The goal of the project is to address this.”
According to data collected by Alachua County, there were 57 traffic-related fatalities in 2017, a 33 percent increase from 2016. These fatalities include “not only vehicle drivers, but also bicyclists and pedestrians.”
Ranka and his team hope to create an open-source analytics solution to enable new transportation applications that utilize data from low-cost video sensors. By creating open-source analytics and using low-cost sensors, the researchers hope to create a system that can be easily replicated in other cities. Data at intersections will be analyzed to monitor the behavior of pedestrians, drivers, and cyclists on the road. The project aligns with the city’s Vision Zero plan, a traffic safety initiative that aims to eliminate all traffic fatalities and severe injuries.
By understanding the risk profile of an intersection through automated detection of near-miss events, communities will be able to proactively design and alter streets and intersections to be safer. Near-miss events are defined by the National Safety Council as unplanned events that did not result in injury, illness, or damage – but had the potential to do so.
“Our technology will lead to less congested and safer traffic systems in Gainesville and other cities,” Ranka said.