Dallas, TX - Khaled Abdelghany, a distinguished professor of civil and environmental engineering at Southern Methodist University (SMU), has clinched a significant three-year, $1.2 million grant from the Federal Highway Administration. The grant is poised to fuel the development of a groundbreaking computer program that leverages artificial intelligence to elevate safety and efficiency at intersections, catering to both vehicles and pedestrians.

The lead researcher, Abdelghany, who is also a fellow at the Stephanie and Hunter Hunt Institute for Engineering and Humanity, spearheads this initiative at the SMU Lyle School of Engineering's Department of Civil and Environmental Engineering. Collaborating on this ambitious project are Professor Michael Hunter from the Georgia Institute of Technology, Director of the Georgia Transportation Institute, and Assistant Professor Mahdi Khodayar from The University of Tulsa.

This venture falls under the umbrella of the Federal Highway Administration's Exploratory Advanced Research (EAR) Program, an initiative collaborating with universities, private companies, and public entities to pioneer transformative research. The EAR Program's ultimate goal is to harness the power of artificial intelligence (AI) and machine learning to enhance the safety and efficiency of transportation systems.

The focal point of the research is traffic intersections, critical junctures influencing highway safety and efficiency. According to the Federal Highway Administration, approximately one-quarter of traffic fatalities and half of all traffic injuries in the United States are attributed to intersections each year.

Improving Traffic Safety at Intersections with AI

Abdelghany, Hunter, and Khodayar are at the forefront of developing PANORAMA: An Interpretable Context-Aware AI Framework for Intersection Detection and Signal Optimization. This innovative program is designed to be applicable to traffic lights at intersections nationwide.

Traditionally, traffic lights operate based on the detection of vehicles approaching an intersection and historical traffic patterns. However, this method falls short in accounting for short-term variations in traffic patterns caused by factors such as weather changes and fails to consider other intersection users like pedestrians, cyclists, and wheelchair users.

Using video cameras, PANORAMA identifies traffic at intersections, categorizing various entities like vehicles and scooters. It then intelligently determines whether the traffic light should display green or red, explained Abdelghany. The team is working on creating an adaptive real-time control system.

PANORAMA seamlessly integrates computer vision technology and optimal control to generate real-time timing plans. These plans enhance intersection safety and efficiency by considering factors such as time of day, weather conditions, and traffic characteristics.

"Ensuring safety for all users is essential for equitable transportation," noted Hunter. "Moreover, PANORAMA will be cost-effective as it does not require additional infrastructure beyond what is already present at many intersections."

Crucially, PANORAMA will implement interpretable AI, providing explanations for its recommendations regarding green or red-light signals, according to Khodayar. This transparency is vital, as it offers essential information to traffic light controller operators, ensuring that the system is not a "black box."

Leveraging SMU's high-performance computing capabilities, the research team will utilize substantial data to effectively train the AI model. Once adequately trained and validated, PANORAMA will be capable of running on any computer.

Not only is PANORAMA expected to make intersections safer and traffic flow smoother, reducing emissions from idling cars, but it will also enable the assessment of each intersection's performance. "We'll be able to assess the performance of each intersection, knowing which ones are operating efficiently and which aren't," emphasized Abdelghany. This groundbreaking research signifies a significant stride towards a smarter and safer transportation future. 

Copyright @ MOTORTIMES, All rights reserved. Do not reproduce without permission.