AI could reduce fatalities at railway crossings

1 min read

Artificial intelligence is being employed in the US to detect trespassing on railroad crossings, an advance that could curb fatalities from train strikes.

AI Detected Grade Crossing Violator Going Around Gates
AI Detected Grade Crossing Violator Going Around Gates - Asim Zaman/ Xiang Liu

Asim Zaman, a Rutgers University project engineer, and Xiang Liu, an associate professor in transportation engineering at the Rutgers School of Engineering in New Jersey, USA created an AI-aided framework that automatically detects railroad trespassing events, differentiates types of violators and generates video clips of infractions. The system uses an object detection algorithm to process video data into a single dataset.

“With this information we can answer numerous questions, like what time of day do people trespass the most, and do people go around the gates when they are coming down or going up?” Zaman said in a statement.

In their research, Zaman and Liu define trespassers as unauthorised people or vehicles in an area of railroad or transit property not intended for public use, or those who enter a signalised grade crossing after it has been activated.

Until now, most research into railroad trespassing was derived from casualty information. But the research overlooked near-misses – occasions Zaman and Liu said can provide valuable insights into trespassing behaviours, which in turn can help with the design of more effective control measures.

To test their theory, the researchers accessed video footage captured at one crossing in urban New Jersey. 

Zaman and Liu trained their AI and deep-learning tool to analyse 1,632 hours of archival video footage from the study site. During 68 days of monitoring, 3,004 instances of trespassing occurred. The researchers also found that nearly 70 per cent of trespassers were men, roughly a third trespassed before the train passed and most violations occurred on Saturdays around 5 p.m. The results are published in the journal Accident Analysis & Prevention.

Zaman said granular data like this could be used by local authorities to position police officers near crossings during periods of peak violations or to inform railway owners and decision makers of more effective crossing solutions – such as grade crossing elimination systems or advanced gates and signals.

“We want to give the railroad industry and decision makers tools to harness the untapped potential of video surveillance infrastructure through the risk analysis of their data feeds in specific locations,” said liu.

In addition to the New Jersey location, the researchers are conducting studies in Virginia and North Carolina and were recently awarded a $583,000 grant from the US Department of Transportation to expand to other states including Connecticut, Massachusetts, and Louisiana.