<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Research in Progress (RIP)</title>
    <link>https://rip.trb.org/</link>
    <atom:link href="https://rip.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Research in Progress (RIP)</title>
      <url>https://rip.trb.org/Images/PageHeader-wTitle-RIP.jpg</url>
      <link>https://rip.trb.org/</link>
    </image>
    <item>
      <title>Adaptive Cyber Threat Detection for Rail SCADA Systems: A Hybrid Machine Learning and Statistical Approach</title>
      <link>https://rip.trb.org/View/2655702</link>
      <description><![CDATA[Supervisory Control and Data Acquisition (SCADA) systems form the digital backbone of modern railway operations, enabling real-time monitoring of critical track geometry parameters including gage, cross-level, alignment, and warp that are essential for preventing derailments and ensuring passenger safety. While SCADA-driven sensing has advanced continuous condition monitoring, it has also introduced new cyber-physical vulnerabilities, particularly stealthy False Data Injection Attacks (FDIAs) capable of masking real defects or fabricating false positives without detection.

Existing rule-based and signature-based detection systems fail to identify subtle or novel attacks in high-dimensional, noisy rail geometry data, and most current models require labeled attack datasets that are rarely available. Although unsupervised methods such as autoencoders and Variational Autoencoders (VAEs) can detect deviations from learned normal behavior, they remain limited by non-stationary data characteristics and static detection thresholds.

This research proposes a Hybrid VAE with Median Absolute Deviation (MAD) scoring to enable robust, adaptive anomaly detection based on the statistical significance of reconstruction errors. The study investigates whether this approach enhances detection of both subtle and overt FDIAs compared to Isolation Forest and static-threshold VAE baselines, evaluates the effectiveness of MAD-based adaptive thresholding against fixed percentile methods, and examines trade-offs in interpretability, computational load, and reliability across attack intensities.

Using an operational track geometry dataset (18,290 samples, 87 features) from Colorado rail testing, the methodology simulates FDIAs through additive spikes, multiplicative distortion, and high-variance noise injection on safety-critical features. Model performance is evaluated using precision, recall, F1-score, and accuracy, with PCA and t-SNE visualization for validation. Findings will provide actionable deployment guidelines for enhancing cyber-physical resilience in railway SCADA systems.]]></description>
      <pubDate>Mon, 19 Jan 2026 16:16:59 GMT</pubDate>
      <guid>https://rip.trb.org/View/2655702</guid>
    </item>
    <item>
      <title>Multi-Sensory System for Railway Track Defect Detection </title>
      <link>https://rip.trb.org/View/2646942</link>
      <description><![CDATA[Railway transportation is essential for moving passengers and freight across the U.S., but accidents continue to pose serious safety and economic risks. In 2022 alone, there were about 950 rail-related fatalities and 6,400 injuries nationwide. While human error and reckless behavior are major contributors, defective track infrastructure is a significant and preventable cause of accidents. Railway tracks are complex systems consisting of steel rails, crossties, fasteners, and ballast, all subject to heavy loads, temperature fluctuations, and environmental impacts. These stresses lead to issues such as broken rails, cracked or spalled crossties, loose or missing fasteners, geometry defects, and cross-level variations. Extreme weather conditions can further cause rail buckling or fracture. Failures in these components can trigger derailments, collisions, hazardous material spills, and major service disruptions. Although manual inspections and specialized vehicles are used, many defects go undetected between inspection cycles. Traditional manual inspections, although reliable for identifying visible rail defects, are labor-intensive and limited in scalability. To improve efficiency, various nondestructive testing (NDT) technologies, such as infrared imaging, acoustic emission, ultrasonic, and electromagnetic techniques, have been used primarily for internal defects. As surface defects become more prevalent, various methods have also been developed for detecting surface-level flaws, which can be broadly categorized into three approaches: static monitoring where sensors at fixed locations provide localized coverage; inspection trolleys which integrate sensors generally in the laboratory setting; and onboard sensing systems which enable real-time detection ahead of moving trains but suffer from high cost with varying imaging quality under different weather and lighting conditions. The primary objective of this project is to develop a comprehensive but low-cost multi-sensory system for railway track defect detection. The system will integrate binocular stereovision cameras, Global Navigation Satellite System / Global Positioning System (GNSS/GPS), and IMU sensors. The scope of this project includes development of a multi-sensory system including controller and field data acquisition, development of real-time data fusion and detection algorithms, and recommendations for system deployment on railway tracks. ]]></description>
      <pubDate>Mon, 05 Jan 2026 23:04:12 GMT</pubDate>
      <guid>https://rip.trb.org/View/2646942</guid>
    </item>
    <item>
      <title>Evaluating Ballast Performance with Freeze/Thaw Cycles</title>
      <link>https://rip.trb.org/View/2573189</link>
      <description><![CDATA[In seasonally cold regions, railroad tracks are subjected to ice formation under sub-freezing conditions and ice thawing under above-freezing conditions due to significant seasonal temperature fluctuations, posing challenges for the maintenance of ballasted railway tracks and operation safety. Currently, little attention has been given to the impact of ice formation and thawing on the permanent deformation of railroad ballast and incidents due to track stiffness variation have not been reported. This proposed research project will investigate the effect of ice formation and thawing on the permanent deformation of ballast through large-scale triaxial cyclic testing, utilizing a newly developed freezing system to simulate frozen conditions. The results will demonstrate the potential track support variation when ballast is subject to freeze-thaw cycles, under the same loading cycles. The rate of permanent deformation will be related to track settlement and help predict track geometry degradation and optimize track maintenance for enhanced track safety.]]></description>
      <pubDate>Mon, 14 Jul 2025 20:12:51 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573189</guid>
    </item>
    <item>
      <title>Rapid Detection of Track Changes from In-Motion Data Acquisition Records: Lab Setup and Field Implementation – Year 3
</title>
      <link>https://rip.trb.org/View/2573191</link>
      <description><![CDATA[Track stiffness is a critical parameter influencing infrastructure integrity, safety, and maintenance efficiency. Track stiffness variations over time and space lead to uneven load distribution, track degradation, and increased risk of failure, necessitating continuous monitoring and timely intervention. Current technologies determine stiffness under loaded or unloaded conditions at discrete locations, or through continuous measurements. They are either costly, labor-intensive, or limited in spatial and temporal resolution. The proposed work is a four-year effort to develop an in-motion system that detects track stiffness and stiffness changes in real-time that is free of the shortcomings of existing techniques. The proposed system is an acceleration-based system that uses hybrid signal processing techniques and machine learning for classification. The system consists of three modules: (1) Data acquisition using onboard vibration sensors; (2) Hybrid signal processing on the edge for feature identification and data compression; and (3) Classification and decision support, utilizing machine learning algorithms for characterization of track conditions in predictive maintenance. This proposal is for Year 3 of the research team's current University Transportation Center for Railway Safety (UTCRS) sponsored effort. Year 1 focused on the development of a track stiffness monitoring concept and produced a feasibility study that led to Year 2 work on method development, and validation through simulations and laboratory small-scale testing. Spurred by the findings of Years 1&2, this proposal focuses on the development of an experimental prototype system and its validation through high-fidelity laboratory testing. In addition, the team proposes to develop a digital twin of the experimental prototype to facilitate extensive validation, calibration, and sensitivity studies to enhance accuracy and scalability. The project will enhance track safety, reduce maintenance costs, and improve railway infrastructure reliability by enabling continuous, cost-effective, and scalable monitoring. The research directly aligns with UTCRS’s strategic goals by advancing infrastructure monitoring technologies and contributes to the United States Department of Transportation (USDOT)’s objectives in safety and economic competitiveness.]]></description>
      <pubDate>Mon, 14 Jul 2025 20:04:39 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573191</guid>
    </item>
    <item>
      <title>Modeling Special Cases of Longitudinal Resistance in Continuously Welded Rail (CWR)</title>
      <link>https://rip.trb.org/View/2573194</link>
      <description><![CDATA[Continuously welded rail (CWR) is the standard for North American freight railroads due to its advantages in ride quality, fatigue life, and reduced maintenance costs, despite concerns about rail buckling and breaks. Longitudinal rail resistance is a critical parameter for re-establishing rail neutral temperature (RNT) after rail breaks and for mitigating potential rail failures caused by vehicle loading, temperature changes, and maintenance activities. This proposed research builds upon a previous year project and continues the effort to refine and enhance the Finite Element (FE) modeling of rail longitudinal resistance. Specifically, it aims to improve the representation of realistic rail and anchor conditions by integrating new experimental data into the FE models. The research will develop efficient 2D and 3D FE models in ABAQUS that incorporate rail-to-tie friction, anchor slip forces, and tie-to-ballast restraint, using both experimental results (e.g., anchor slip behavior under varying load conditions) and historical data (e.g., rail-sleeper friction and sleeper-ballast resistance). The models will accommodate various rail profiles, tie materials, and geometric configurations, and will be applicable to a wide range of track conditions including frozen ballast, frozen structures, turnouts, crossings, and loading scenarios from vehicles and maintenance activities. The proposed project will be executed through four key interconnected areas of research: (1) Effects of sleeper-ballast on models larger than 4-ft in length using FE modeling in ABAQUS, (2) experimental testing in the laboratory for anchor slippage with various anchor types, (3) sensitivity analysis, and (4) model analysis with various track conditions. ]]></description>
      <pubDate>Mon, 14 Jul 2025 19:49:53 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573194</guid>
    </item>
    <item>
      <title>Temperature-Induced Cyclic Loading Effects on Rail Anchor Slip Force</title>
      <link>https://rip.trb.org/View/2573195</link>
      <description><![CDATA[Recently, continuous welded rail (CWR) systems have been widely adopted due to their enhanced ride quality, reduced maintenance requirements, and extended service life for both rails and rolling stock. However, the elimination of joints in CWR introduces challenges, particularly in managing thermal expansion, which can lead to track buckling. A critical factor in maintaining track stability is the Rail Neutral Temperature (RNT) — the temperature at which rails are free of thermal stress. Anchors, which resist longitudinal rail movement, play a key role in managing RNT and ensuring track integrity. While previous studies have largely focused on the static behavior of rail anchors, this research emphasizes the importance of cyclic longitudinal loading, which can simulate daily and seasonal temperature fluctuations. Unlike static loading, cyclic longitudinal loading on the rail-anchor under different temperatures can potentially lead to gradual degradation in anchor performance, slip initiation, or cumulative displacement over time. These effects may be more critical to track stability than static forces alone, especially under service operating conditions. This study will conduct full-scale laboratory testing to investigate the impact of cyclic temperature-induced longitudinal loading on slip force performance for various rail anchor types. By simulating temperature cycles and measuring anchor slip under controlled conditions — including different anchor geometries, installation tightness, and environmental parameters — this research aims to provide an understanding of the long-term reliability of rail anchoring systems under thermal cycling. Also, this study addresses the need to construct a 15-foot full-scale track segment on ballast and wood ties to replicate in-field conditions for the future studies to be performed for this project.]]></description>
      <pubDate>Mon, 14 Jul 2025 19:42:30 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573195</guid>
    </item>
    <item>
      <title>Intelligent Aerial Drones for Railroad Track Traversability Assessment, Intrusion Detection 
and Integrity Evaluation</title>
      <link>https://rip.trb.org/View/2573854</link>
      <description><![CDATA[Aerial drones have been increasingly used in railroad operations as they offer an effective low-cost solution that can be easily deployed and efficiently support the human efforts in inspection and monitoring activities. This proposal outlines the development of an advanced system leveraging intelligent aerial drones for comprehensive railroad track monitoring and evaluation. The project serves as the integration phase (phase 3) of two University Transportation Center for Railway Safety (UTCRS) projects that in the previous two phases developed related technology: (i) a project on the development of Intelligent Aerial Drones for Traversability Assessment of Railroad Tracks, and (ii) a project on the development of AI-enabled system for Track Intrusion Detection and Track Integrity Evaluation. Through this integration, an intelligent aerial drone will be developed able to carry equipment for the autonomous inspection of railroad tracks with the following capabilities: (i) Visual-based identification and autonomous following of the track; the system will be able to work even in GPS-degraded environments (tunnels, dense forests); (ii) Collision avoidance capability where the drone senses and avoids obstacles; (iii) Track centering capability where the drone follows the same line regardless of the number of tracks in the field of view; (iv) Identification and mapping of any obstacles identified blocking the line; (v)Intrusion\Trespassing detection; and (vi) AI-based Detection, Classification, Tracking, and Situational Evaluation. This innovative solution promises to improve operational efficiency, safety, and cost-effectiveness in the management of railroad networks, while minimizing downtime and enhancing system reliability.]]></description>
      <pubDate>Mon, 14 Jul 2025 19:12:30 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573854</guid>
    </item>
    <item>
      <title>Experimental Determination of Rail Fracture Properties</title>
      <link>https://rip.trb.org/View/2573188</link>
      <description><![CDATA[There are approximately 1,100 train derailments per year in the U.S., and rail fracture is responsible for 7% of these derailments. Fatigue cracking is a widespread rail fracture issue and a significant safety concern accompanied by critical rail maintenance costs. Despite this fact, a reliable model for predicting fatigue fracture in rails has not yet been deployed within the U.S. In this UTCRS project, the research team has developed an advanced computational algorithm for predicting crack evolution in rails subjected to cyclic fatigue loading. The team's fracture model demonstrates the feasibility and scientific rigor over the traditional phenomenological approaches, while several challenges remain for its successful practical implementation. One of the core challenges is to identify fracture properties in the model when rails are subjected to long-term cyclic fatigue loadings. This project aims to determine the fracture properties of railheads subjected to long-term cyclic fatigue loading. Toward that end, the will use will use its nonlinear cohesive zone (NCZ) fracture model implemented within the finite element computational algorithm and experimental results from the railhead fatigue testing which is currently under development and will be continued in 2025-2026 (as described in our companion UTCRS proposal entitled Experimental Determination of Crack Growth in Rails Subjected to Long-Term Cyclic Fatigue Loading). Successful identification of rail fracture properties (i.e., fundamental material properties) through this project will serve as a core piece for the development of TAMU’s rail fracture modeling framework which will significantly impact the current railway safety and asset management program. This project will be carried out with direct interaction and supervision by MxV Rail personnel.]]></description>
      <pubDate>Mon, 14 Jul 2025 13:04:59 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573188</guid>
    </item>
    <item>
      <title>Advanced Model for Predicting Buckling in Rails</title>
      <link>https://rip.trb.org/View/2573186</link>
      <description><![CDATA[It is well-known that track buckling is one of the most commonplace causes of train
derailments. Accordingly, with partial funding provided by the research team's previous US Department of Transportation (DOT) University Transportation Center (UTC) and the Technology Transportation Center, Inc. (TTCI, now MxV Rail), the team is continuing to develop a track buckling model for deployment by MxV Rail as a tool for predicting track buckling, A significant advancement over currently deployed track buckling models, the team's technology includes an open-source nonlinear finite element algorithm that is user-friendly. Briefly, the team's track buckling model accounts for the effects of the following on track buckling: both longitudinal and lateral track walk; rail neutral temperature (RNT); both lateral and longitudinal cross tie aggregate interfacial friction; track modulus; nonlinear track liftoff; and broken spikes. In addition, it is sufficiently robust to be capable for additional environmental causes to be described herein and in a companion proposal. Given these advanced capabilities, track engineers will be able to dramatically improve track safety.]]></description>
      <pubDate>Mon, 14 Jul 2025 13:01:29 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573186</guid>
    </item>
    <item>
      <title>Computational Model for Predicting Fracture in Rails Subjected to Long-Term Cyclic Fatigue Loading</title>
      <link>https://rip.trb.org/View/2573185</link>
      <description><![CDATA[It is well known that one of the most significant causes of train derailments within the U.S. is due to rail fracture. Despite this fact, a reliable model for predicting fatigue fracture in rails has not yet been deployed within the U.S. The research team has recently been developing an advanced computational algorithm for predicting crack evolution in ductile solids subjected to long-term cyclic loading. In this UTCRS project, the research team will continue to adapt this model to the prediction of crack growth in rails. Concomitantly, with funding provided by MxV Rail, the research team has recently completed a decade-long series of experiments designed to provide data usable for the purpose of developing just such a model. The research team, therefore, possesses the ability to both predict crack growth due to cyclic fatigue in rails, as well as to utilize our previously obtained experimental results to validate our predictive methodology. Hence, the research team has begun the following rather challenging task of: 1) modifying our computational model for predicting crack growth for application to cyclic fatigue in rails; 2) developing an experimental protocol for obtaining the material properties required to deploy their computational fracture model (described in their companion project entitled Experimental Determination of Crack Growth in Rails Subjected to Long-Term Cyclic Fatigue Loading); 3) demonstrate the effectiveness of their model for predicting the effects of long-term cyclic loading on rail fracture; and 4) develop a procedure based on their model for railway engineers to utilize to determine when rails should be inspected and potentially removed from service for cause, thereby increasing rail safety. This project will be carried out with direct interaction and supervision by MxV Rail engineers. ]]></description>
      <pubDate>Mon, 14 Jul 2025 12:45:49 GMT</pubDate>
      <guid>https://rip.trb.org/View/2573185</guid>
    </item>
    <item>
      <title>Large-scale Testing for Detecting Changes in Track Modulus with Low-cost Sensors Installed on Rolling Stock



</title>
      <link>https://rip.trb.org/View/2572334</link>
      <description><![CDATA[U.S. railroads transport 1.6 billion tons of freight over more than 140,000 miles of track each year. Safe and efficient operation of such a vast infrastructure requires extensive monitoring, evaluation, and maintenance of its track systems. Track modulus is a critical parameter in the design, analysis, and maintenance of railroad track, as it is indication of  the material stiffness below the rail comprising the combined stiffnesses per unit length of rail, plates, ties, ballast, and subgrade. Locations of “soft” track can cause increases in rail deflections and stresses, which increases the rate of rail deterioration. Traditionally, track modulus is measured onsite using static deflection testing where a known load is applied to the track, the resulting deflection is measured, and this deflection is extrapolated to the track modulus. However, this is time consuming and labor-intensive, especially if measurements are to be taken at multiple points along the track. To address the challenges with track modulus measurement, this research attempts to refine existing methods and expand the scale of monitoring by leveraging data from railcars to identify problem locations on the track and relate measured responses to specific track deficiencies. The approach allows making continuous estimations over long sections of the track and also is less costly, as self-contained acceleration and data acquisition systems are inexpensive and easy to attach to the vehicle bodies. This research will use technologies such as the ground penetrating radar (GPR) and track geometry cars in combination with low-cost sensors placed on the existing plant of rolling stock for accelerated track monitoring. Track response to track conditions will be measured. Constitutive load-deflection relationships will be established between track condition, loading, and deflection to determine the track modulus. The modulus will be characterized by leveraging the mechanics that relates vehicle accelerations to the condition of track upon which the instrumented vehicle travels. The research team is experienced in the use of low-cost accelerometers for bridge monitoring and assessment, where it established the mechanistic relationships between loading and response under various conditions and showed the feasibility of determining quantifiable, specific conditions from the acceleration data. Using this expertise and experience, the team now seeks to develop mechanics-based relationships that correlate railcar body acceleration profiles to track behavior and, eventually, track condition. This project is the next step towards full implementation, where it will use previously characterized conditions on a model test to support the constitutive load-deflection relationships. The overall impact of this work will be the widespread monitoring of track infrastructure through the installation of low-cost accelerometers on existing rollingstock. The industrial partner, BNSF, will provide support and help with implementation.]]></description>
      <pubDate>Wed, 09 Jul 2025 16:08:22 GMT</pubDate>
      <guid>https://rip.trb.org/View/2572334</guid>
    </item>
    <item>
      <title>UAV-Imagery Based Track Component Health Condition Inspection</title>
      <link>https://rip.trb.org/View/2572335</link>
      <description><![CDATA[This project will develop a unmanned aerial vehicle (UAV)-imagery based intelligent track component health condition inspection system, that will utilize a camera and global positioning system (GPS) in a UAV integrated with edge computer device to identify missing and broken fasteners at the real-time speed. There are a considerable number of studies on the use of UAVs in track inspection. However, these studies utilize drones as a carrier of cameras and need human pilots to operate and control the drones. The collected images are stored onboard for a later analysis at some centralized facility. So, the current practices based on these studies has several drawbacks and limitations. Human pilot cost can be significant. Images collected by different pilots at the same track segment could vary and depend on the pilots’ operation skill, experience, and judgement. The inspection route is also subjective. The delay between data collection, data processing, and decision making depreciate the value of the inspections because track conditions can quickly deteriorate as traffic accumulates. To address these limitations, this project proposes a next generation UAV-imagery based track inspection system featuring advanced computer vision for real-time fattener defect defection and  efficient edge computing for field data processing. The advanced, embedded computer vision model will extract the features of various track components to evaluate their health conditions, such as missing or broken spikes, clips, rail surface detect, welding crack, broken ties. All inspected data will be immediately processed onboard for track condition assessment without the need for intensive data storage or transferring. The processed results will also be linked to the image-acquisition locations with the on-board GPS unit of the UAV. Both software and hardware are based on a modular and open-source design, which makes it compatible and transferable to other drone platforms, which to the best of the proposer’s knowledge is still unavailable in commercial or academic sectors. The proposed research consists of three modules: Module I --Training Image Library module; Module II -- Artificial Intelligence (AI)-based Track Component Detection module; and Module III -- Edge-computing system module. Module I will establish specialized drone-based track image database for convolutional neural network (CNN)-based computer vision model training. In Module II, a pixel-level detection system will be developed by using a tailored instance segmentation model to detect track components in a fast and accurate fashion. In Module III, to enable in-situ image analysis and AI inference, an appropriate mobile edge-computing platform and integration strategy will be developed. The proposed system will significantly reduce inspection cost and derailment risk, optimize maintenance strategy, and improve track safety. It will also greatly reduce the workload and improve the work conditions for the track inspectors because the system will automatically process the images and record the detected defects and access the track where it is hard to reach for the inspectors.]]></description>
      <pubDate>Wed, 09 Jul 2025 16:04:48 GMT</pubDate>
      <guid>https://rip.trb.org/View/2572335</guid>
    </item>
    <item>
      <title>Innovations Deserving Exploratory Analysis--The Transit IDEA Program. IDEA 107. Rail Track Geometry Estimation via Fusion of Inertia and Vision Data with Track Deflection Compensation</title>
      <link>https://rip.trb.org/View/2572331</link>
      <description><![CDATA[The United States has the world's longest railway network serving both freight and passenger transportation. For safe and reliable operation of the trains, railway tracks are critically important as the train stability, speed, and ride quality hinge upon them. Poor track geometry can cause  enhanced wear and tear on the train components, increase maintenance costs, and raise risk of derailments. For this reason, railway tracks are surveyed regularly to ensure that they meet the required safety standards. To assist in doing this, track geometry and inspection vehicles (TGIVs) provide alignment and profile information. However, inherent inaccuracies of TGIV estimates leads to increased reliance on manual surveys. The high cost of surveys discourages the needed frequent track geometry monitoring and assessment. 

This research project proposes to integrate camera modules with TGIVs and fuse image data with Inertial Measurement Unit (IMU) measurements to provide cost effective and faster and more accurate railway track geometry estimates and condition insights. IMU measurements are excellent for monitoring high frequency position and orientation variations in tracks while camera-based measurements are great for observing low frequency variations. So, by fusing image data with IMU measurements, it is possible to produce highly accurate and complete track geometry data. 

The project work will be carried out in collaboration with the New Jersey Transit, the partner transit agency. Specific geometric parameters required by the partner transit agency as well as the agency’s  hardware (e.g., track and TGIV) specifications will be determined. Image processing and feature tracking algorithms for vision-based data extraction will be developed along with deflection compensation algorithm based on example TGIV data provided to the research team. A data fusion algorithm will then be developed, based on state estimator, AI, and statistical approaches. All algorithms will be combined into an automated framework, and their capabilities tested using computer simulation. A camera module with the required specifications will be procured, installed, and tested to identify the ideal module placement on the TGIV with respect to position and orientation. The camera will be installed on a TGIV and several passes over an off-line track will be conducted to compare its results with the laser scans of the track being tested. 

The proposed approach has a high potential for replacing the need for separate TGIV and surveying deployments and provide a unified and continuous geometry estimation platform that would improve safety and  reliability of railroads at a reduced cost for operators.]]></description>
      <pubDate>Tue, 08 Jul 2025 17:04:18 GMT</pubDate>
      <guid>https://rip.trb.org/View/2572331</guid>
    </item>
    <item>
      <title>Local Resonances-based NDE Technique for Rail Flaw Detection</title>
      <link>https://rip.trb.org/View/2572328</link>
      <description><![CDATA[Rail internal defects have been one of the leading causes of track-related accidents. Rail internal defects can reduce cross-sectional area and introduce stress concentration. Moreover, they can develop with normal, rapid, and sudden growth rates. If left undetected, internal defects can result in broken rails, train accidents, and derailments, where sudden rail rupture can occur without warning. Accurate and reliable rail flaw detection is therefore critically important for improving safety and reliability and minimizing the risks of accidents induced by rail internal defects. Nondestructive evaluation (NDE) techniques, including roller search unit (RSU), ultrasound A-Scan, and phased array, have been employed to detect rail internal defects but their performance or accessibility has been limited. This project will develop a new technology using a newly identified wave propagation phenomenon  (local resonances in rails) for rail defect detection.  These local resonances feature highly localized energy and signature frequencies that are governed by the geometry and material properties of a rail. These local resonances were found to be sensitive to internal defects over the full rail cross-section and are easy to measure. A low-cost contactless acoustic sensing prototype will be developed that would generate local resonances in rails. These resonances will provide flaw detection capability over the full rail section. The prototype’s sensing configuration will be simple and robust and, compared with existing NDE techniques, it will not require sophisticated/expensive sensors or data acquisition systems. It will combine fast data collection with efficient data processing to produce timely critical damage alert. If successful, the project is expected to have a significant impact of the current state of practice for accuracy and practicality with regard to determining the presence and severity of internal rail defects.

 ]]></description>
      <pubDate>Tue, 08 Jul 2025 16:46:41 GMT</pubDate>
      <guid>https://rip.trb.org/View/2572328</guid>
    </item>
    <item>
      <title>Topological Data Analysis and Track Geometry Data</title>
      <link>https://rip.trb.org/View/2446876</link>
      <description><![CDATA[Rail geometry defects constitute a major cause of accidents in the United States. Geometry related accidents are often very severe and damaging. While rail geometry-caused derailments continue to increase according to Federal Railroad Administration (FRA) safety data, track quality analysis remains effectively unchanged. The use of TQI or track quality index takes a narrow view of track assessment by focusing on quality without considering safety. The bipartite analysis of track quality and safety results into two maintenance types: routine and corrective maintenance respectively. This report shows how to create a hybrid index that combines both element of safety and geometry quality to predict only one maintenance regime based on track condition. It is an initial step towards the big picture of creating indices that will be iterated based on maintenance savings and defect probability thresholds. This study employs a linear and nonlinear dimension reduction technique that expresses the probability distribution of observations based on the similarity or dissimilarity in their embedded space whilst also maximizing the variance in data. This study found application in principal component analysis (PCA) and T-Stochastic neighbor embedding (TSNE) for separating geometry defects from higher dimensional space to lower dimensions. Results show that while both techniques effectively reduces track geometry data, PCA yields a potential defect probability threshold in spite of TSNE being a better geometry defect predictor.
This study employs a linear and nonlinear dimension reduction technique that expresses the probability distribution of observations based on the similarity or dissimilarity in their embedded space whilst also maximizing the variance in data.
]]></description>
      <pubDate>Tue, 29 Oct 2024 15:25:20 GMT</pubDate>
      <guid>https://rip.trb.org/View/2446876</guid>
    </item>
  </channel>
</rss>