A collaboration between Marshall University’s College of Engineering and Computer Sciences and the Marshall Advanced Manufacturing Center (MAMC) is helping bring cutting-edge technology to the railroad industry. Teams developed and manufactured a high-speed railroad track scanning system for the U.S. Army Corps of Engineers that uses cameras, sensors and artificial intelligence to detect rail defects faster, safer and more accurately than traditional inspections.
The project began more than two years ago when Dr. Ammar Alzarrad, associate professor in Marshall’s Department of Civil Engineering, presented the concept to the Engineer Research and Development Center (ERDC) in Mississippi, the research arm of the U.S. Army Corps of Engineers. Impressed by the proposal, ERDC awarded funding to support development of the technology.
Traditionally, railroad inspections rely heavily on visual observation. Inspectors manually examine tracks for defects, a process that can be time-consuming and prone to human error. The new scanning system uses depth-of-field sensors, cameras and artificial intelligence to identify defects more quickly and accurately.
The scanner can detect issues such as split rails, missing bolts and other defects that affect rail performance. The system currently recognizes five types of defects, with additional capabilities expected as the software continues to improve. When a defect is identified, the system generates a geotagged image pinpointing the exact location for repairs.
The technology could significantly reduce disruptions to rail operations. The U.S. Army Corps of Engineers inspects approximately 1,500 miles of railroad track each year, often requiring sections be shut down during inspections. Because the new scanner can operate while moving, inspections can be performed more efficiently with less interruption to rail traffic.
Mounted to a Ford pickup truck, the scanner operates at speeds up to 35 miles per hour, with a project goal of eventually reaching 40 miles per hour. Maintaining stability at those speeds presented one of the project’s biggest engineering challenges.
“We do testing from five miles per hour all the way up to 35 miles per hour,” Alzarrad said. “The slower you are, the higher the accuracy. The faster you are, the more vibration and the more problems. But the ask from the Army Corps of Engineers is to go up to 40 miles per hour.”
The first prototype was constructed from wood by Marshall engineering faculty, students and lab technicians. While effective as an initial proof of concept, the unit was too heavy and unstable to produce reliable scans at higher speeds.
To solve the problem, the College of Engineering team designed a new lightweight aluminum cart and partnered with MAMC to manufacture it.
“MAMC’s skilled people made this happen from machined aluminum,” Alzarrad said. “Now it is lightweight, more customizable and easier to adjust. We can extend the arms holding the cameras to cover both sides of the railroad or change the configuration based on our needs.”
MAMC mechanical engineers, machinists and technicians manufactured the metal cart, which attaches to a high-rail truck and supports five cameras — two on each side and one overhead — to provide a comprehensive view of the tracks.
Alzarrad said the railroad scanner project is a great example of how different university departments collaborate to innovate.
“It means the world that we trust MAMC, and we know they will be able to deliver,” he explained. “They were fast, they were accurate. It was a very smooth and easy process. We are Marshall. They are Marshall, and we work together. It wasn’t hard.”
May 28, 2026
