Track Surface Irregularity Position Localization with Smartphone-Based Solution

Tracks are a critical and expensive railroad asset, requiring frequent maintenance. Railroads companies often rely on the accurate localization and identification of the track anomalies that could cause serious damage to infrastructure, environment, and the traveling public. However, the deployed method of inspection and maintenance is expensive, slow, increases the safety risk of workers, and requires track closure. Also, the technical limitations of present methods prevent their network-wide scaling to all railroads. Low-cost global positioning system (GPS) receivers and accelerometers aboard regular vehicles offer a promising alternative to monitor all railroads in real time. However, low resolution and low accuracy of GPS receivers and non-uniform sample rates of the inertial sensors produced signal position misalignment and additive signal noise. Subsequently, signal-to-noise ratio (SNR) decreases and the detection error to locate the track irregularity will increase. In particular, false positives and false negatives can increase when SNR decreases. The introduced framework in this research reduces detection error while enhancing the quality of the signals and extracted features. This research demonstrates the potential use of low-cost sensors aboard hi-rail vehicles to monitor automatically and continuously for inertial events caused by irregular track geometry. The study characterizes and validates its accuracy by comparing the estimated positions of detected irregularities with the actual positions of irregularities that the railroad inspector observes. Therefore, railroad agencies that employ developed frameworks and methods will benefit from reliable track and equipment condition situation to make informed decisions, leading to resource optimization.

• Record URL:
  • https://www.ugpti.org/resources/reports/details.php?id=1049
• Record URL:
  • https://rosap.ntl.bts.gov/view/dot/59905
• Supplemental Notes:
  • This document was sponsored by the U.S. Department of Transportation, University Transportation Centers Program.
• Corporate Authors:

  North Dakota State University, Fargo

  Transportation, Logistics and Finance Department
  Fargo, ND  United States 

  Mountain-Plains Consortium

  North Dakota State University
  Fargo, ND  United States  58108

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Authors:
  • Lu, Pan
  • 0000-0002-1640-3598
  • Bridgelall, Raj
  • 0000-0003-3743-6652
  • Tolliver, Denver
  • 0000-0002-8522-9394
  • Bhardwaj, Bhavana
  • 0000-0002-4379-1565
  • Dhingra, Neeraj
  • 0000-0001-9970-7185
• Publication Date: 2021-11

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Report
• Features: Figures; Photos; References; Tables;
• Pagination: 51p

Subject/Index Terms

• TRT Terms: Flaw detection; Maintenance of way; Railroad tracks; Sensors; Signal processing; Smartphones
• Subject Areas: Data and Information Technology; Maintenance and Preservation; Railroads;

Filing Info

• Accession Number: 01831012
• Record Type: Publication
• Report/Paper Numbers: MPC-21-445
• Contract Numbers: MPC-551
• Files: UTC, NTL, TRIS, ATRI, USDOT
• Created Date: Dec 21 2021 9:44AM