Aircraft Manoeuvre Classification Using Adaptive Time Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately identify and classify aircraft manoeuvres due to varying durations and require advanced data processing, complicating the analysis of time trends.
Innovation Solution
A method and system that utilizes sensors to monitor aircraft parameters, constructs a training data structure, extracts feature vectors, and employs multiple classifiers to detect and classify manoeuvres using time windows and statistical analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If advanced data processing techniques are used to identify manoeuvres from time trends, then manoeuvre identification accuracy is improved, but system complexity increases
Solution Approach 1:
The patent segments the continuous time trends into discrete manoeuvre events by identifying start and end points based on threshold criteria. This segmentation transforms the complex continuous analysis into simpler discrete event detection, resolving the contradiction between accuracy and complexity.
Solution Approach 2:
The patent extracts specific feature parameters from the comprehensive time trends, focusing only on the most discriminative features for manoeuvre identification. This extraction reduces the data processing burden while maintaining identification accuracy by eliminating redundant information.
2Adaptability or versatility
If different manoeuvres with different durations are analyzed using time trends, then comprehensive manoeuvre coverage is improved, but analysis difficulty increases
Solution Approach 1:
The patent employs dynamic time window adjustment that adapts to different manoeuvre durations. The analysis window automatically extends or contracts based on the detected manoeuvre type, enabling comprehensive coverage of various manoeuvre durations while simplifying the analysis through adaptive parameter adjustment.
Solution Approach 2:
The patent changes the analysis parameters (time window size, sampling rate, threshold values) according to the specific manoeuvre being analyzed. This parameter adaptation allows the system to handle diverse manoeuvre types effectively without requiring complex fixed-structure analysis for each case.
Data Source
AI summary
Method implemented through a computer for detecting the execution, by an aircraft, of a manoeuvre belonging to a macrocategory among a plurality of macrocategories, including: receiving a data structure with a plurality of time series of values of quantities relating to a flight of the aircraft; for each time duration among a plurality of predetermined time durations, selecting a corresponding subset of the data structure and extracting a corresponding feature vector; on the basis of the feature vectors, generating a corresponding input macrovector and applying to the input macrovector an output classifier to generate estimates indicative of the probability that the aircraft was performing manoeuvres belonging to the macrocategories.