3D Component Scanning for Railway Vehicle Defect Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for inspecting vehicle components in motion, such as trains, are hazardous and time-consuming due to reliance on manual visual checks, which are not reproducible and lack precision in detecting small defects or mispositioning of components like screws or bolts.

Innovation Solution

A detection system using scanners with lasers and cameras generates three-dimensional representations of vehicle components, aligns them with reference models, and calculates rigid transformations to automatically detect bad positioning and surface defects, even for components as small as millimeters, eliminating the need for manual intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If manual visual inspection is used to check vehicle components, then the inspection can be performed with simple equipment, but the inspection is hazardous, time-consuming, and not reproducible

Engineering Contradiction:
Improveinspection equipmentVSAvoidinspection speed
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent replaces manual visual inspection with an automated optical scanning system using lasers and cameras. The scanning system captures three-dimensional representations of vehicle components in motion, eliminating the need for manual inspection while improving productivity and eliminating hazards associated with workers approaching moving vehicles.

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

Solution Approach 2:

The system enables self-inspection of vehicle components through automated scanning and processing. The computer automatically aligns three-dimensional representations with reference models, detects defects, and generates reports without human intervention during the inspection process, making the system self-sufficient and highly productive.

Inventive Principle:
Principle #25Self-service

2Device complexity

If manual visual inspection is used to check vehicle components, then the equipment required is simple, but the detection precision for small components like screws or bolts is insufficient

Engineering Contradiction:
Improveinspection equipmentVSAvoiddefect detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces manual visual inspection with an automated optical scanning system using lasers and cameras. The scanning system captures three-dimensional representations of vehicle components in motion, eliminating the need for manual inspection while improving productivity and eliminating hazards associated with workers approaching moving vehicles.

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

Solution Approach 2:

The patent transitions from two-dimensional visual inspection to three-dimensional scanning and representation. By capturing and processing three-dimensional data of components, the system achieves superior detection precision for small components like screws or bolts, as the additional spatial dimension enables more accurate measurement and comparison against reference models.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Extent of automation

If alignment with reference models is used to detect component positioning, then automated detection is achieved, but the precision is not sufficient to detect millimeter-sized components or surface defects

Engineering Contradiction:
Improvedetection automationVSAvoidcomponent positioning accuracy
Core Design Contradiction:
Extent of automationVSMeasurement precision

Solution Approach 1:

The patent transitions from two-dimensional visual inspection to three-dimensional scanning and representation. By capturing and processing three-dimensional data of components, the system achieves superior detection precision for small components like screws or bolts, as the additional spatial dimension enables more accurate measurement and comparison against reference models.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent creates accurate three-dimensional digital copies (representations) of vehicle components through laser scanning. These digital copies are then aligned with and compared to reference models, enabling precise automated detection of positioning errors and surface defects. The copying process preserves fine geometric details that can be analyzed with high precision computationally.

Inventive Principle:
Principle #26Copying

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

This system enables accurate, reproducible, and rapid detection of component mispositioning and defects, reducing the risk of human error and time consumption, while allowing for precise identification of small parts like bolts or screws, improving safety and efficiency.

Implementation Method 1

a scanning system (18) for scanning the specific components of the vehicle (1) in motion, comprising at least one laser and at least one camera

Methodology Applied
Scientific EffectLaser: Laser

Data Source

PatentEP2998927B1Method for detecting the bad positioning and the surface defects of specific components and associated detection device
Publication Date: 2018.09.05 ALSTOM TRANSPORT TECH SAS
  • EP2998927B1 patent drawingFigure 1
  • EP2998927B1 patent drawingFigure 2
  • EP2998927B1 patent drawingFigure 3~4

AI summary

A method for detecting the bad positioning and the surface defects of a specific component of a vehicle in motion, comprising: - scanning the region surrounding a specific component of a vehicle in motion on a rail road to obtain a three dimensional representation of the region surrounding the specific component, - aligning the three dimensional representation with a surrounding three dimensional model of the region surrounding the specific component to obtain a roughly aligned three dimensional representation, characterized in that, the method also comprises : - aligning the roughly aligned three dimensional representation with a specific three dimensional model of the specific component to obtain an accurately aligned three dimensional representation and an accurate rigid transformation, and - analyzing the bad positioning and the surface defects of the specific component.