Investigation of GPR-Based Self-Localization for Rail Vehicles: Evaluating Track Structures and a Correlation-Based Approach
Noll M, Kohnert S, Caldero P, Seidel C (2025)
Publication Type: Journal article
Publication year: 2025
Journal
DOI: 10.1109/OJIM.2025.3619319
Abstract
In this study, an approach to self-localization for rail vehicles using ground-penetrating radar (GPR) is evaluated. This method involves matching a current measurement window with a prerecorded map. Typical substructure and superstructure materials of railway lines are analyzed, and their effects on GPR imaging are discussed. An initial preprocessing chain is introduced, and the properties of processed GPR scans are examined. Three specific measurement properties that reduce the comparability of preprocessed measurement windows, thus hindering matching, are identified and explained. To address these challenges, a normalization-based correlation approach is proposed, which scales the data to counteract disruptive properties. Its performance is evaluated by its matching success rate over a 24km route with four track types, including ballasted and ballastless tracks. Within a 10km search range, successful matches were possible in 53.8% of the cases for ballastless tracks with sound absorbers and 99.4% for mixed ballastless and ballasted tracks.
Involved external institutions
Germany (DE)How to cite
APA:
Noll, M., Kohnert, S., Caldero, P., & Seidel, C. (2025). Investigation of GPR-Based Self-Localization for Rail Vehicles: Evaluating Track Structures and a Correlation-Based Approach. IEEE Open Journal of Instrumentation and Measurement. https://doi.org/10.1109/OJIM.2025.3619319
MLA:
Noll, Maximilian, et al. "Investigation of GPR-Based Self-Localization for Rail Vehicles: Evaluating Track Structures and a Correlation-Based Approach." IEEE Open Journal of Instrumentation and Measurement (2025).
BibTeX: Download