Aircraft System Testing with Classifier-Based Operation Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aircraft system testing methods are inefficient in identifying and addressing differences between desired and actual operations, particularly due to the complexity of aircraft systems and the limitations of existing testing methods that may not account for user preferences and real-world scenarios, leading to costly revisions after system deployment.
Innovation Solution
A test apparatus and method using a trained classifier to compare actual aircraft system operations with a modeled operation, generating signals for differences, and allowing for continuous training based on user feedback and data from various sources, including aircraft users, to adapt to changing preferences and operational conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional aircraft system testing methods are used, then testing can be performed with existing tools, but the ability to identify differences between desired and actual operations is insufficient
Solution Approach 1:
The patent introduces a trained classifier as an intermediary component that mediates between the aircraft system under test and the testing apparatus. The classifier models the desired operation of the aircraft system and compares it with actual operation, enabling precise identification of operational differences without requiring complex manual testing procedures
Solution Approach 2:
The patent replaces traditional mechanical and manual testing methods with an automated computational system. The testing apparatus uses software-based classification algorithms and automated data processing to substitute for manual testing procedures, thereby improving measurement precision while managing system complexity through automation
2Reliability
If comprehensive testing of aircraft systems is performed, then operational differences can be identified, but development time and testing costs increase
Solution Approach 1:
The patent applies preliminary action by training the classifier in advance using training data that represents desired aircraft system operations. This pre-training phase allows the classifier to be ready for immediate use in testing, enabling rapid comparison and identification of operational differences without requiring extensive testing time for model preparation
Solution Approach 2:
The patent creates a computational copy of the desired aircraft system operation through the trained classifier. This virtual model replicates the expected behavior of the aircraft system, allowing for rapid comparison with actual operations and enabling comprehensive reliability testing without requiring multiple physical prototypes or extended development cycles
3Adaptability or versatility
If the classifier is continuously trained with user feedback, then the model adapts to changing preferences and conditions, but the complexity of maintaining the testing system increases
Solution Approach 1:
The patent implements a feedback mechanism where user inputs and mobile device data are continuously collected and used to retrain the classifier. This feedback loop enables the model to adapt to changing user preferences and operational conditions, improving adaptability while the automated nature of the feedback collection and processing manages the complexity of maintaining the training system
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is test apparatus (100) for testing at least part of an aircraft system. The test apparatus (100) is configured to obtain first data (210) comprising a plurality of input values and process the input values using a classifier (230) configured to model an operation of the at least part of an aircraft system (220). Second data is obtained, comprising a set of output values (240b). The output values (240b) from the second data are processed with output values (240a) generated by the classifier to identify differences between operation of the at least part of an aircraft system (220) and the operation as modelled by the classifier (230).