Rail Asset Survey System Using Multi-Sensor Fusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional rail track asset surveying methods are time-consuming, costly, and inefficient, relying heavily on manual processes and slow data capture, which leads to outdated information and significant delays in identifying and assessing asset conditions, especially due to the computational intensity of video image processing and handling large volumes of image data.

Innovation Solution

A railroad track asset surveying system utilizing a combination of image sensors, laser measurement sensors, and asset classifiers that generate two or three-dimensional data sets for automated asset classification and status assessment, allowing for real-time data processing and analysis, including the use of shape, surface property, brightness, color, and texture sensors, and thermal imaging, to provide immediate insights into asset conditions and operational risks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual inspection methods are used with inspectors walking along the track or traveling on slow speed platforms, then asset data can be recorded with human knowledge and judgment, but the surveying process becomes extremely time-consuming and disrupts normal track usage

Engineering Contradiction:
Improveasset identification accuracyVSAvoidsurveying time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical inspection with an automated optical-mechanical system consisting of cameras, lasers, and processors that capture and analyze track asset data automatically, eliminating the need for human inspectors to physically traverse the tracks while maintaining accurate asset identification through automated recognition algorithms

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service inspection where the automated surveying equipment independently captures, processes, and analyzes track asset data without requiring human intervention during the surveying process, allowing continuous operation at normal track speeds while maintaining reliable asset identification

Inventive Principle:
Principle #25Self-service

2Quantity of substance

If conventional video image processing with off-site computer analysis is used, then large volumes of image data can be captured and reviewed, but the processing becomes computationally intensive and delays action deduction

Engineering Contradiction:
Improveimage data volumeVSAvoiddata processing time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent segments the data processing function by distributing computational tasks between on-board processors that perform initial analysis of image and laser data during data capture, and off-site computers that handle comprehensive analysis of the structured results, enabling parallel processing that reduces overall computation time while maintaining thorough asset assessment

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary data processing and structuring on-board during the surveying operation, organizing raw image and laser data into standardized formats with embedded metadata before transmission, which prepares the data for rapid off-site analysis and enables immediate action deduction without waiting for complete data transfer and initial processing

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system enables rapid, automated, and accurate surveying of rail track assets, reducing the time and cost associated with manual inspections, providing real-time asset identification, classification, and condition assessment, thereby improving operational efficiency and safety by minimizing delays between data capture and actionable insights.

Implementation Method 1

image sensors... to record video of the track and its surroundings

Methodology Applied
Scientific EffectLight reflection: Reflection

Implementation Method 2

laser measurement sensors... which can provide a two or three dimensional view of the scene

Methodology Applied
Scientific EffectLIDAR (Light Detection and Ranging): LIDAR

Implementation Method 3

thermal image capture sensor

Methodology Applied
Scientific EffectThermal radiation: Thermal Radiation

Data Source

PatentEP3138754B1Rail track asset survey system
Publication Date: 2024.03.27 RAIL VISION EURO
  • EP3138754B1 patent drawingFigure 1
  • EP3138754B1 patent drawingFigure 2
  • EP3138754B1 patent drawingFigure 3

AI summary

There is provided a railroad track asset surveying system suitable for mounting to a conventional passenger or freight vehicle. Image capture sensors including linescan, areascan and laser imaging devices are mounted on a railroad vehicle. A location determining system, such as GPS and/or a track position sensor, is provided for images captured by the image capture sensor. An image processor provides an asset classifier for detecting assets in captured images and classifying the detected assets by assigning an asset type to the detected asset from a predetermined list of asset types according to one or more feature in the captured image. The image processor also provides an asset status analyser for identifying an asset status characteristic and comparing the identified status characteristic to a predetermined asset characteristic so as to evaluate a deviation therefrom. The system may automatically survey a variety of assets, including vegetation, in a scene surrounding the rail track. The system may be provided as a singular assembly or module for mounting in a forward facing direction to a conventional vehicle.