Aircraft Onboard Prediction via Distributed Edge Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high volume of data generated by aircraft engines poses a challenge for centralized data management and processing, making it impractical to predict future changes in operation by comparing current states with previous states across a large fleet, as it requires significant infrastructure and processing power.
Innovation Solution
A method is implemented on board each aircraft to calculate its current state, request similarity analysis with previous states of similar aircraft, and analyze changes in operation, distributing processing and storage while reducing communication volume by storing and analyzing data locally and sharing only relevant information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If centralized data management and processing is implemented to predict future changes in aircraft operation by comparing current states with previous states across a large fleet, then prediction accuracy is improved, but infrastructure complexity and processing requirements increase significantly
Solution Approach 1:
The patent divides the centralized prediction system into distributed edge computing nodes deployed on individual aircraft and ground stations. Each node independently performs local data processing and prediction tasks, segmenting the monolithic centralized system into multiple autonomous units that collectively achieve fleet-wide prediction capabilities without requiring a single complex centralized infrastructure
Solution Approach 2:
The patent transitions from a single centralized processing dimension to a multi-dimensional distributed architecture spanning airborne edge devices, ground-based servers, and cloud platforms. This dimensional expansion allows prediction workloads to be distributed across multiple layers, reducing the complexity burden on any single infrastructure component while maintaining overall prediction accuracy
2Loss of information
If all aircraft data is transferred to ground for processing, then comprehensive analysis is improved, but communication volume and data transfer requirements increase significantly
Solution Approach 1:
The patent extracts critical prediction algorithms and processing capabilities from the ground-based centralized system and deploys them directly on airborne edge devices. This extraction enables aircraft to perform comprehensive local data analysis and generate predictions in-flight, eliminating the need to transfer all raw data to ground while preserving complete analytical capabilities
Solution Approach 2:
The patent implements preliminary data processing, filtering, and feature extraction at the edge devices before data leaves the aircraft. By performing these actions in advance, the system reduces the volume of data requiring ground communication to only essential processed results and anomalies, while maintaining the ability to conduct comprehensive analysis through pre-deployed algorithms
Data Source
AI summary
A method of estimating future change in operation of a monitored aircraft (A), including the following steps performed by a computer on board the monitored aircraft, to calculate a current state of the monitored aircraft (ECA) from measurements (VFA) of variables related to operation of the monitored aircraft, to send a request to analyze the similarity of the calculated current state with previous states (EPB) of similar aircraft, and to analyze change in operation (SPB) of each similar aircraft having a similar previous state to determine a probable change (FPA) in operation of the monitored aircraft. The method includes a step performed by a computer (B) on board each aircraft, to compare the calculated current state (ECA) with previous states (EPB) of similar aircraft, and send the change in operation (SPB) corresponding to an identified similar previous state. The invention includes an on board system capable of implementing the method.