ASU AI research drives road safety innovations

Fulton Schools faculty members create new tech and trainings to assist transportation planners

For many of us, driving is a daily routine. But it’s also one of the most dangerous things we do. Road traffic collisions remain the leading cause of death for children and young adults around the world. As autonomous vehicles cruise into the mainstream, Arizona State University researchers are hitting the accelerator on finding safer ways to share the road.

In the School of Computing and Augmented Intelligence, part of the Ira A. Fulton Schools of Engineering at ASU, “YZ” Yezhou Yang is shifting gears with innovative research that could steer us toward a safer future.

Yang is a Fulton Schools associate professor of computer science and engineering and a thought leader in the development of a type of artificial intelligence, or AI, known as computer vision. In this specialty of computer science, machines use sensors, cameras and complex software programs to perceive the world around them. The technology fuels everything from self-driving cars to smart robots. Now, thanks to Yang’s latest development, it’s ready to take on traffic safety.

Yang’s team has created ETram, a groundbreaking event-based traffic monitoring system that’s revving up the way we collect and use road data. ETram doesn’t just idle. It records up to 10 hours of detailed information on lighting conditions, weather and traffic events.

But unlike traditional traffic cameras, ETram keeps personal privacy in the rearview mirror. Instead of capturing identifiable images, it collects only data.

“The cities of the future must address concerns about personal privacy and security,” Yang says. “ETram records data and not images, providing information that can help train AI models but doesn’t use anyone’s image without their knowledge.”

The research has some real traction. Cameras around campus are already capturing data to help spot near misses and high-risk areas, especially after dark, when driving gets dicey.

Next, the high-speed team, including Fulton Schools Assistant Teaching Professor Bharatesh Chakravarthi, is building machine learning models that will turbocharge ETram’s impact. Insights provided by the system can be the roadmap city planners need to design safer intersections and smarter streets. Armed with helpful information, transportation planners can adjust traffic lights, install signage and improve the design of new intersections.

Workshop attendees listen to a presentation in Abu Dhabi. At a two-day event, Fulton Schools faculty members led transportation officials through hands-on exercises designed to show how AI can help improve traffic planning and management. Photo courtesy of the International Road Federation

Taking new tech on the road

Because making sure city planners and officials have access to this new technology is a critical part of the mission, Kamil Kaloush is leading efforts to raise awareness of how AI can be used to improve traffic safety.

Kaloush is the FORTA Professor of Pavement Engineering in the School of Sustainable Engineering and the Built Environment, part of the Fulton Schools. He currently serves as the elected chairman of the board of the International Road Federation, or IRF Global, where he works to implement the organization’s vision in more than 70 countries.

In that role, Kaloush collaborated with other Fulton Schools faculty members, including Yang and Hua Wei, an assistant professor of computer science and engineering, to develop a workshop for transportation planners.

“ASU is an IRF Global university member, and we support the organization with various requests for training and benchmarking,” Kaloush says. “Our mission is to advance the transfer of knowledge among road and transportation professionals and experts, ensuring the development of safe and sustainable infrastructure.”

In April, the team test drove the workshop, “AI and Big Data Applications for Future Traffic and Incident Management,” at an event in the United Arab Emirates hosted by Abu Dhabi Mobility and IRF Global.

Organized and coordinated by Aliaksandr Smirnou, director of IRF Global for the Middle East and North Africa, the two-day workshop served as a touchpoint for cross-sector knowledge exchange, drawing participation from government agencies, transportation authorities, researchers and private sector tech leaders.

Mohammad Farhadi Bajestani, a Fulton Schools adjunct faculty member, led in-person sessions while the rest of the team participated remotely. Sessions focused on harnessing AI and real-time analytics to tackle congestion, improve emergency response and lay the groundwork for autonomous vehicle solutions.

Attendees also worked hands-on with AI technology, learning more about how they could develop and train their own models. Those tutorials were designed to get experts comfortable with creating customized tools.

The ASU team is working with IRF Global to offer the workshop in additional locations in the future. Kaloush says that the long-standing partnership with IRF Global has well positioned the university to offer such trainings. Discussions are also underway with the city of Phoenix for a similar event to help apply this technology in our own neighborhoods.

Yang says that with efforts like ETram and the transportation workshops, his team is on the road to making streets safer for all.

“Now it’s back to work. No pit stops!” he says with a laugh.

With Fulton Schools faculty members leading the convoy, the future of transportation safety isn’t on the horizon. It’s coming at full throttle.