Case Study

"Bihrle’s computer vision capabilities have been used in conjunction with our railway safety enhancement research and the FAA’s Pathfinder Program. The breadth of railway anomaly detection capabilities provided by Bihrle allows us to further research into the use of long range UAS."

Todd Graetz

Director, Technology Services at BNSF


In 2014, BNSF Railway selected Ardenna's parent company, Bihrle Applied Research Inc., as its key consultant in pursuing its UAS initiatives. Since 2014, Bihrle has provided support to BNSF through a wide range of activities that included engineering consulting, UAS regulatory guidance, development of a UAS simulator for drone crew training, and the development of UAS flight planning software.


One of the key enabling technologies developed for BNSF was RailVisionTM, a computer-vision based software solution designed to automatically process the terabytes of imagery being collected by drones during BVLOS supplemental railway inspection flights. RailVisionTM automatically performs the detection, classification and reporting of features and anomalies in order to generate actionable reports in a fraction of the time required by traditional methods.

Business Case

Traditional Methods

Drones & RailVision

When it comes to supplemental railroad track inspections, today’s railway companies are faced with several limitations:


  • Inspection equipment requires track occupancy
  • Single track inspection only
  • Limited actionable information
  • Significant delay (months) in delivery of results

The use of drones and RailVision™ provides a unique solution at an unequaled value to the rail transportation industry.


  • No track occupancy
  • Inspect multiple tracks simultaneously
  • Wide range of automated detection abilities
  • Delivery of actionable results in hours

Features & Capabilities

RailVisionTM provides the capability to inspect

hundreds of miles of track, assessing multiple tracks simultaneously and covering a wide range of detections including:


  • Multiple tracks simultaneously
  • Rail breaks and gap detection
  • Crosstie slew, spacing, and count
  • Crosstie type and condition
  • Fastener type and count
  • Frog detection
  • Spike/hole pattern
  • Switch detection and position
  • Ballast fouling

The RailVision Process

Demonstrated Performance


81 Research Drive | Hampton, VA 23666

© 2019. Ardenna. All Rights Reserved.