The National Academies

Rail Safety IDEA Project 43 [Completed (IDEA)]

Augmenting Reality for Safer Inspections of Railroad Infrastructure and Operations

  Project Data
Funds: $100,000
Staff Responsibility: Inam Jawed
Research Agency: University of New Mexico
Principal Investigator: Fernando Moreu
Fiscal Year: 2020

This research project explored the use of Augmented Reality (AR) to enhance the accuracy and efficiency of monitoring railroad inspections and operations. The proposed approach included modeling human cognition and learning processes and programming them into AR. Specific AR applications were developed for data collection during railroad inspections. This report describes the AR software, the laboratory experiments, the field tests, and the professional/scientific collaborations performed to evaluate the human factors and opportunities to explore the quantification of human cognition with railroad activities using AR. A new AR application to quantify the inspection quality, named AR for railroad attention using Eye tracking (ARRA-Eye) technology was developed. This ARRA-Eye application was designed to measure and quantify the inspector’s attention during inspection of the railroad as well as other surfaces and operations such as cracks on rail or ties. Other AR applications developed to explore the interface between the inspector and the railroad included “concrete crack”, “rail crack documenting and deployment”, “rail gauge measurer”, “AR robot control”, and “AR sensing.” The validation of the AR applications included testing by Class I railroad staff at the University of New Mexico (UNM) campus as well as a visit to Canadian National (CN) Railway headquarters in Chicago, Illinois. Other railroad interactions included collaboration with high-speed railroad engineers from Japan who visited UNM, AR testing with expert railroaders at the American Railroad Engineering and Maintenance-of-Way Association (AREMA) meeting in Denver, Colorado, and field inspections of concreate cracks with the New Mexico Department of Transportation. The interaction with industry early in the project informed the updates to the AR programming proposed for both training of new inspectors as well as its potential use for data collection in the field at moderate distances (between three and nine feet) where AR can be useful for distance measurement without the need of ladders or when it is safer to measure these distances without contact.

The final report is avaliable. 

To create a link to this page, use this URL: http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=5081