DigiOnTrack
Digital On Track
The DigiOnTrack project is developing a cost-effective, wireless monitoring system for tracks and vehicles, based on simple retrofit solutions. In the long term, the project aims to reduce maintenance costs for rail infrastructure and vehicles through condition-based maintenance planning. The sub-project lays the foundation for the implementation of preventive and predictive maintenance strategies. The system is intended to be capable of detecting potential problems before they occur and recommending preventive measures to minimize downtime and repair costs.
- Funders Federal Ministry for Digital Affairs and Transport
- Funding program mFUND
- Funding amount 0.78 Million Euro
- Duration 2023–2025
Funding
Partners
Team
- Prof. Dr.-Ing. Jan-Niklas Voigt-Antons
Project Lead - Michael Stern
Research Scientist - Michelle Hallmann
Research Scientist
Intro / Project Overview
DigiOnTrack is a collaborative research project funded by the German Federal Ministry for Digital and Transport (BMDV) through the mFUND program. The consortium brings together industrial and academic partners, including 5micron GmbH, Ostakon GmbH, DESAG Deutsche Eisenbahn Service AG, and Hamm-Lippstadt University of Applied Sciences. Running from 2023 to 2025 with a total project volume of 2.4 million euros, DigiOnTrack aims to develop an intelligent predictive maintenance system for rural rail infrastructure. By combining structure-borne noise sensors, secure data transfer technologies, and machine-learning-based event classification, the project seeks to increase reliability, safety, and operational efficiency across regional rail networks.Main Goal & What the System Does
The DigiOnTrack project is designed to transform how railway infrastructure health is monitored and how early-warning information reaches drivers and maintenance teams. Traditionally, infrastructure faults are detected through periodic inspections or manual observations, which often leave subtle wear patterns unnoticed until they escalate. DigiOnTrack addresses this by equipping trains with low-power vibration and acoustic sensors that continuously record data during regular service. These measurements are analyzed in a centralized backend that detects anomalies, classifies events, and provides structured insights to drivers, dispatchers, and workshop personnel. Through the combination of sensor data, GPS positioning, and user-generated annotations, the system enhances situational awareness and enables timely maintenance interventions.Innovation & Research Contribution
A key innovation of DigiOnTrack lies in its user-centered approach to predictive maintenance. Train drivers and workshop foremen actively contribute annotations and feedback, which serve as high-quality training data for supervised machine-learning models. This participatory labeling process ensures that real-world operational knowledge directly informs algorithmic decision-making. Additionally, digital dashboards provide intuitive access to historical data, detected anomalies, and sensor health, supporting informed decision-making across the entire maintenance chain. The project also employs tamper-resistant data storage mechanisms to ensure transparency and traceability. Through iterative testing, prototyping, and data-driven refinement, DigiOnTrack establishes a scalable and trustworthy approach for future predictive maintenance systems in regional rail transportation.Publications
- Dishonesty Tendencies in Testing Scenarios Among Students with Virtual Reality and Computer-Mediated Technology
Kojić, T., Vergari, M., Dovhalevska, A., Möller, S. & Voigt-Antons, J.-N. (2024, June). Dishonesty Tendencies in Testing Scenarios Among Students with Virtual Reality and Computer-Mediated Technology. Paper presented at the International Conference on Human-Computer Interaction (HCII 2024). Washington DC, USA. https://doi.org/10.1007/978-3-031-61953-3_14 - Optimized User Experience for Labeling Systems for Predictive Maintenance Applications
Hallmann, M., Stern, M., Vona, F., Franke, U., Ostertag, T., Schlüter, B. & Voigt-Antons, J.-N. (2024, June). Optimized User Experience for Labeling Systems for Predictive Maintenance Applications. Paper presented at the International Conference on Human-Computer Interaction (HCII 2024). Washington DC, USA. https://doi.org/10.1007/978-3-031-76821-7_4 - Sales Skill Training in Virtual Reality: An evaluation utilizing CAVE and Virtual Avatars
Vona, F., Stern, M., Ashrafi, N., Schorlemmer, J., Stemann, J. & Voigt-Antons, J.-N. (2025, June). Sales Skill Training in Virtual Reality: An evaluation utilizing CAVE and Virtual Avatars. Paper presented at the International Conference on Human-Computer Interaction (HCII 2025). Gothenburg, Sweden.
