Aircraft Inspection Data Validation Before UAS Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aircraft inspection methods using unmanned aircraft systems (UAS) lack automated validation of captured data, leading to variability in quality and untrustworthy results, which can result in undetected anomalies and non-compliance with regulatory requirements.
Innovation Solution
A computer system that receives captured data from UAS, compares it with reference data, and determines whether it is within a set of tolerances for valid data. If the data is invalid, the system performs corrective actions before detecting anomalies, ensuring the data meets quality and regulatory standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated inspection systems are used without data validation, then inspection speed increases, but data quality and reliability deteriorate
Solution Approach 1:
The system performs preliminary validation of captured data against reference data before the data is used for anomaly detection. This preliminary action ensures that only valid data within acceptable tolerances proceeds to analysis, preventing unreliable data from compromising inspection results while maintaining automated inspection speed
Solution Approach 2:
The system establishes a feedback loop where captured data is continuously compared against reference data and tolerances. When data falls outside acceptable ranges, the system generates feedback signals to trigger corrective actions such as recapturing data or adjusting inspection parameters, thereby maintaining data quality without manual intervention
2Reliability
If manual evaluation of captured images is performed, then data quality control improves, but inspection time and costs increase
Solution Approach 1:
The system enables self-service quality control by automatically validating captured data against pre-established reference data and tolerances. The automated inspection system performs its own quality assessment without requiring manual evaluation, eliminating the need for specialized inspectors while maintaining consistent quality standards across all inspections
Solution Approach 2:
The system replaces manual mechanical evaluation processes with automated computational validation. Instead of human inspectors visually examining images, the system uses computer-based algorithms to compare captured data against reference data, substituting human labor with automated processing that is both faster and equally reliable
3Measurement precision
If automated data validation is implemented, then inspection accuracy improves, but system complexity increases
Solution Approach 1:
The system performs preliminary validation of captured data against reference data and predefined tolerances before anomaly detection. This preliminary filtering ensures that only valid data within acceptable ranges proceeds to analysis, improving inspection accuracy by preventing false conclusions from invalid data while maintaining automated operation
Solution Approach 2:
The system uses parameter-based validation by comparing captured data parameters against reference parameters and predefined tolerance ranges. This parameter-driven approach enables automated accuracy control through numerical comparisons rather than complex qualitative assessments, improving precision while keeping the validation logic relatively simple
Data Source
AI summary
A method, apparatus, system, and computer program product for inspecting an aircraft. A computer system receives captured data associated with a flight path flown by an unmanned aircraft system to acquire the captured data for the aircraft. The computer system compares the captured data with reference data for the aircraft to form a comparison. The computer system determines whether the captured data is within a set of tolerances for valid captured data using a result of the comparison. Prior to detecting anomalies for the aircraft using the captured data, the computer system determines a set of corrective actions in response to the captured data being outside of the set of tolerances for the valid captured data in which the set of corrective actions is performed.