News Release

SMU, research partners using artificial intelligence to make traffic intersections safer, more efficient

$1.2 million Federal Highway Administration grant to SMU, Georgia Tech and University of Tulsa to fund 3-year study

Grant and Award Announcement

Southern Methodist University

Rain simulation

image: 

PANORAMA integrates computer vision technology and optimal control to create real-time timing plans that enhance intersection safety and efficiency, taking into account various factors such as time of day, weather conditions like rain, and traffic characteristics.

view more 

Credit: SMU (Southern Methodist University)

SMU (DALLAS) – Khaled Abdelghany, a professor of civil and environmental engineering at SMU (Southern Methodist University), has been awarded a three-year, $1.2 million grant by the Federal Highway Administration. The grant aims to develop a computer program that utilizes artificial intelligence to enhance the safety and efficiency of intersections for both vehicles and pedestrians.

The federal funding was granted to the lead researcher, Abdelghany, who is a professor at the SMU Lyle School of Engineering's Department of Civil and Environmental Engineering and is a fellow at the Stephanie and Hunter Hunt Institute for Engineering and Humanity. Professor Michael Hunter at the Georgia Institute of Technology and Director of the Georgia Transportation Institute and Assistant Professor Mahdi Khodayar at The University of Tulsa are co-researchers on the grant.

This grant is part of the Federal Highway Administration's Exploratory Advanced Research (EAR) Program, which collaborates with universities, private companies, and public entities conducting pioneering research in these areas. The EAR Program's goal is to leverage artificial intelligence (AI) and machine-learning technology to make transportation safer and more efficient.

Traffic intersections are key to highway safety and efficiency. Each year, roughly one–quarter of traffic fatalities and about one–half of all traffic injuries in the United States are attributed to intersections, the Federal Highway Administration reports.  

Improving traffic safety at intersections with AI

Abdelghany, Hunter and Khodayar are developing a program called PANORAMA: An Interpretable Context-Aware AI Framework for Intersection Detection and Signal Optimization. This program can be applied to traffic lights at intersections throughout the country.

Typically, traffic lights at intersections are programmed to switch between red and green based on detection of vehicles as they approach an intersection and historic traffic patterns. However, this approach does not adequately consider short-term variations in the traffic pattern due to things like changes in the weather, and it does not account for other intersection users, such as pedestrians, cyclists, and wheelchair users.

"Using video cameras, PANORAMA will identify the traffic at these intersections, categorizing vehicles, scooters, or any other entities present. PANORAMA will then determine whether the traffic light should display green or red," Abdelghany explained. "We are devising an adaptive real-time control system."

PANORAMA integrates computer vision technology and optimal control to create real-time timing plans that enhance intersection safety and efficiency, taking into account various factors such as time of day, weather conditions, and traffic characteristics.

"Ensuring safety for all users, including pedestrians, cyclists, scooter users, and those with disabilities, is essential for equitable transportation. Moreover, PANORAMA will be cost-effective as it does not require infrastructure beyond that already found at many intersections," Hunter noted.

Importantly, PANORAMA will implement what’s known as interpretable AI.

“Utilizing AI PANORAMA will not be a black-box,” Khodayar explained. “Instead, it will provide an explanation for its recommendations regarding green or red-light signals, offering essential information to the traffic light controller operators.”

Given that substantial data is necessary to train AI effectively, the research team will utilize SMU's high-performance computing capabilities to develop the model. However, once the system is adequately trained and validated, PANORAMA will be capable of running on any computer.

Not only will PANORAMA help intersections be safer and traffic run smoother – cutting down on emissions from idling cars – but “we’ll be able to assess the performance of each intersection, knowing which ones are operating efficiently and which aren’t,” Abdelghany said. 

This work is supported by the Federal Highway Administration under Agreement No. 693JJ32350030.

 

About SMU

SMU is the nationally ranked global research university in the dynamic city of Dallas. SMU’s alumni, faculty and more than 12,000 students in eight degree-granting schools demonstrate an entrepreneurial spirit as they lead change in their professions, communities and the world.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.