Skip Navigation

Mid-America Transportation Center

Study of a Distributed Wireless Multi-Sensory Train Approach Detection and Warning System for Improving the Safety of Railroad Workers

Final Report
click to download report


Researchers

  • Principal Investigator: Hamid Sharif (hsharif@unl.edu 402-554-3628)
  • Project Status
    Complete
    About this Project
    Brief Project Description & Background
    This project proposes to research the key aspects of multi-sensory data acquisition and classification for detection of approaching trains in the proximity of track work sites to ensure the safety of railroad track workers. We envision a distributed system that collaboratively processes data and shares its findings to accommodate detection on all tracks in the area. We envision a system that is rechargeable, communicates wirelessly, and will be able to interface with the Positive Train Control (PTC) infrastructure for future-proof operation and integration with other safety aspects like in-cab alerting.
    Research Objective
    This work will focus on the research of sensing systems to detect approaching trains using a variety of techniques such as track circuits, EM field propagation, visual and auditory detection, vibration sensing, etc., and their utilization in a multi-sensory approach to overcome individual method's shortcomings such as limited range, reliability, and detection accuracy. This work will also develop methods for distributed data classification for the accurate detection of trains approaching a work site and alerting of the track workers in the area.
    Potential Benefits
    Currently, the railroads rely on their own personnel stationed a distance away from track work sites to act as lookouts to spot approaching trains. Unfortunately, this is highly error-prone. Humans are simply not suitable for prolonged periods of high levels of alertness and tend to easily become distracted. This has resulting in a growing number fatalities and accidents per year. To ensure the safety of railroad workers, novel methods for data acquisition and classification to automate the detection of approaching trains are needed. Current technology does not provide sufficient warning time, nor the required accuracy and reliability and has led to the railroad industry reliance on human lookouts instead. Our proposed solution is expected to provide a solution to this problem and drastically improve railroad worker safety.
    Abstract
    The Federal Railroad Administration strongly promotes safer railroad operations throughout the Nation's railroad industry. One area of concern is the safety of railroad workers who are often required to work on active mainline tracks or in their close proximity. To prevent accidents, workers have to be aware of approaching trains with enough time to move equipment and personnel to a safe distance from the track.
    Project Amount
    $ 59910