Track Intrusion Detection and Track Integrity Evaluation

Other than train collisions, track intrusion (also referred to as track fouling) is a major factor causing railroad accidents, especially at the railroad-highway crossings. According to the Report to Congress “National Strategy to Prevent Trespassing on Railroad Property” issued by the Federal Railroad Administration (FRA), trespassing is currently the number one cause of all railroad-related deaths. The number of fatalities due to trespassing, including both illegally entering and remaining in the railroad right-of-way is even higher than the number of fatalities due to collision between vehicles and trains. The impact of loss of lives, but also the financial and societal impact associated with those accidents, is enormous. The FRA report indicated that the accidents during 2012 to 2016 have cost $43 billion to the nation. Unfortunately, at present, there is no dedicated system to tackle the issues associated with trespassing or other anomalous situations (e.g., suicide) and enhance railroad safety. Current track intrusion detection relies on high-rail inspection which is labor-intensive and requires significant track time. Clearly, there is an urgent need to develop practical solutions to identify track intrusion and mitigate risks of potential accidents. Railroad crossings are the locations where most of the trespassing has taken place, and almost three quarters of all trespassing events were located within 1000 feet of a crossing. This is largely due to the fact that pedestrians and vehicles alike cross the track through grade crossings. Therefore, it is a higher priority to address trespassing within the grade crossing area. However, it should be noted that the proposed effort is also generally applicable to broader areas along the track that are far away from the crossings. With the development of unmanned aerial vehicles (UAVs), including autonomous UAVs, it is possible to develop an autonomous track intrusion detection and track integrity evaluation system to identify any track fouling conditions ahead of collision and share critical information to both railroads and local first responders in time to minimize loss due to a potential impact. The system will integrate a surveillance unit, real-time communication unit, and computer vision and deep-learning artificial intelligence (AI) unit on an edge computing platform. Furthermore, the proposed system will be integrated to the proposed Intelligent Aerial Drones for Traversability Assessment of Railroad Tracks project. The success of this research will significantly enhance situational awareness at grade crossings or other installation locations, mitigate train collision risk, reduce local law enforcement workload, improve quality of life, and benefit all the stakeholders in industry, railroads, and local, state, and federal administration and legislation.


  • English


  • Status: Active
  • Funding: $78551
  • Contract Numbers:


  • Sponsor Organizations:

    Office of the Assistant Secretary for Research and Technology

    University Transportation Centers Program
    Department of Transportation
    Washington, DC  United States  20590

    University of South Carolina, Columbia

    502 Byrnes Building
    Columbia, SC  United States  29208
  • Managing Organizations:

    University of South Carolina, Columbia

    502 Byrnes Building
    Columbia, SC  United States  29208
  • Project Managers:

    Stearns, Amy

  • Performing Organizations:

    University of South Carolina, Columbia

    502 Byrnes Building
    Columbia, SC  United States  29208
  • Principal Investigators:

    Qian, Yu

    Rizos, Dimitris

    Vitzilaios, Nikolaos

  • Start Date: 20230601
  • Expected Completion Date: 20240831
  • Actual Completion Date: 0
  • USDOT Program: University Transportation Centers Program

Subject/Index Terms

Filing Info

  • Accession Number: 01898082
  • Record Type: Research project
  • Source Agency: University Transportation Center for Railway Safety
  • Contract Numbers: 69A3552348340
  • Files: UTC, RIP
  • Created Date: Oct 31 2023 9:40PM