Near-Real-time Health Monitoring and Assessment of a Railway Track system

Project Abstract

The extensive rail network in the United States serves as a lifeline for the mobility of people and goods, underscoring the critical need for robust methods to evaluate, monitor, and predict the health of rail infrastructure. The primary aim of this research is to create an evaluation framework that enables the rigorous analysis of the efficiency and mobility implications of various transportation policies, especially within the context of the Washington DC area. This research aims to implement a near real-time health monitoring and assessment technique that integrates contact-based strain and temperature sensors, contactless camera image or video data, and innovative machine learning (ML) models to achieve reliable long-term assessment and predictive modeling of railway track systems. This research will be implemented and validated in real-time at selected sections of a rail-track system in the region. The project's outcome will provide a validated method for early warning of railway track systems. This method will provide information to decision makers of rail systems in near-real time to take actions that can reduce delays and accidents by targeting speed advisories at specific locations.

Universities Involved

Howard University

Principle Investigators

Dr. Claudia Marin

Funding Sources and Amounts

USDOT: $100,000

Start Date

September 1, 2023

Completion Date

September 1, 2024

Expected Research Outcomes & Impacts

The project's outcome will provide a validated method for early warning of railway track systems. This method will provide information to decision makers of rail systems in near-real time to take actions that can reduce delays and accidents by targeting speed advisories at specific locations.

Subject Areas

Railways, Safety, Machine Learning