Vehicle-Borne Surface Inspection With Image-Guided Tactile Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional surface inspection methods for large-scale components, such as aircraft fuselages, are time-consuming and prone to inconsistency due to human error, and existing sensors lack the accuracy to detect small defects like scratches and gouges efficiently.
Innovation Solution
A two-stage inspection method using image sensors to predict defects and tactile sensors to confirm and measure them, where image sensors provide initial detection and tactile sensors verify defects with high precision, optimizing coverage and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a sensor with small field of view is used for surface inspection, then measurement precision is improved, but productivity deteriorates due to impracticable sweeping across large surfaces
Solution Approach 1:
The inspection process is divided into two segments: a first sensor (image sensor) performs a first inspection to identify candidate defects, and a second sensor (tactile sensor) performs a second inspection to confirm and characterize them. This segmentation allows the system to efficiently screen large areas while maintaining high measurement precision for confirmed defects.
Solution Approach 2:
The first sensor performs a broader, less precise inspection to identify candidate defects, applying partial action to the entire surface. The second sensor then applies excessive action (more precise measurement) only to the candidate defects identified by the first sensor, rather than inspecting every point on the surface with high precision.
2Productivity
If a sensor with large field of view is used to cover large surfaces, then productivity is improved, but measurement precision deteriorates due to insufficient accuracy for small defects
Solution Approach 1:
The inspection system uses two different sensors with different field of view characteristics. The first sensor (image sensor) has a larger field of view for efficient coverage, while the second sensor (tactile sensor) has a smaller field of view for precise measurement of candidate defects.
Solution Approach 2:
The system replaces a single mechanical scanning approach with a multi-sensor system that combines optical imaging (first sensor) with tactile sensing (second sensor), allowing each sensor to perform its optimized function.
3Productivity
If laser-line systems or structured light scanners are used for defect inspection, then productivity is improved through automated scanning, but measurement precision deteriorates due to inability to measure sharp discontinuities with sufficient accuracy
Solution Approach 1:
The system replaces optical scanning methods (laser-line systems, structured light scanners) with tactile sensing (second sensor) for the confirmation and characterization stage, as tactile sensors can directly contact and measure sharp discontinuities with the required precision.
Data Source
AI summary
The present disclosure provides a method of inspection of a surface of an aerodynamic structure. The method includes acquiring image(s) of the surface using image sensor(s) disposed adjacent to the surface, and determining, using the image(s) applied to a model, predicted defect(s) of the surface and corresponding location information. The method further includes controlling, using the location information, the position of tactile sensor(s) disposed adjacent to the surface to acquire dimensioning information for at least a first predicted defect of the predicted defect(s). The method further includes characterizing the first predicted defect using the dimensioning information.