image:
PANORAMA integrates computer vision technology and optimal control to create real-time timing plans that improve intersection safety and efficiency by taking into account various factors such as time of day, weather conditions such as rain, and traffic characteristics. To do.
view more
Credit: SMU (Southern Methodist University)
SMU (Dallas) – SMU (Southern Methodist University) civil and environmental engineering professor Khaled Abdelghani has been awarded a three-year, $1.2 million grant from the Federal Highway Administration. This grant aims to develop a computer program that utilizes artificial intelligence to make intersections safer and more efficient for both vehicles and pedestrians.
Federal funding was awarded to principal investigator Abdelghani. Mr. Abdelghani is a professor in the Department of Civil and Environmental Engineering in the SMU Lyle School of Engineering and a research associate in the Stephanie Hunter Hunt Institute for the Anthropology of Engineering. Georgia Tech professor Michael Hunter, director of the Georgia Transportation Institute, and University of Tulsa assistant professor Mahdi Hodayal are co-investigators on the grant.
This grant is part of the Federal Highway Administration grant. Exploratory Advanced Research (EAR) Program, we collaborate with universities, private companies, and public organizations that are conducting pioneering research in these areas. The goal of the EAR program is to leverage artificial intelligence (AI) and machine learning technology to make transportation safer and more efficient.
Intersections are key to highway safety and efficiency. According to the Federal Highway Administration, intersections are believed to be responsible for approximately one-quarter of traffic fatalities and one-half of traffic injuries in the United States each year. report.
Improving traffic safety at intersections with AI
Abdelghany, Hunter, and Khodayar are developing a program called PANORAMA (Interpretable Context-Aware AI Framework for Intersection Detection and Signal Optimization). This program is applicable to traffic lights at intersections nationwide.
Intersection traffic lights are typically programmed to switch between red and green based on the detection of vehicles approaching the intersection and historical traffic patterns. However, this approach does not fully account for short-term changes in traffic patterns, such as due to changes in weather, nor does it take into account other intersection users, such as pedestrians, cyclists, and wheelchair users.
“PANORAMA uses video cameras to identify traffic conditions at these intersections and classify vehicles, scooters, and other objects. PANORAMA then uses video cameras to identify whether traffic lights display green or red. ,” Abdelghani 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 improve intersection safety and efficiency by considering various factors such as time of day, weather conditions, and traffic characteristics.
“Ensuring the safety of all users, including pedestrians, cyclists, scooter users, and people with disabilities, is essential to fair transportation.Furthermore, the panorama goes beyond what is already seen at many intersections. It’s cost-effective because it doesn’t require as much infrastructure,” Hunter said. I got it.
Importantly, PANORAMA implements so-called interpretable AI.
“Using AI PANORAMA is not a black box,” Kodayal explained. “Instead, we will provide recommendations for green or red lights and provide important information to traffic light controllers.”
Training AI effectively requires large amounts of data, so the research team plans to leverage SMU’s high-performance computing capabilities to develop the model. However, once the system is properly trained and validated, PANORAMA can be run on any computer.
In addition to improving intersection safety and smoothing traffic by reducing emissions from idling vehicles, PANORAMA also “evaluates the performance of each intersection to determine which intersections are operating efficiently. , we will be able to know which intersections are not,” Abdelghani said. Said.
This work is supported by the Federal Highway Administration under Agreement No. 693JJ32350030.
About SMU
SMU A nationally ranked, world-class research university located in the dynamic city of Dallas. SMU alumni, faculty, and her more than 12,000 students across eight degree-granting schools demonstrate entrepreneurial spirit while leading change in their professions, communities, and the world.
Disclaimer: AAAS and EurekAlert! We are not responsible for the accuracy of news releases posted on EurekAlert! Use of Information by Contributing Institutions or via the EurekAlert System.