Flight Path Uncertainty Quantification for Aerial Vehicle Trajectory Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Four-dimensional flight trajectories generated by aerial vehicle flight management systems are subject to uncertainties such as weather conditions, leading to inaccuracies in trajectory-based operations, which can result in deviations from predicted flight paths, especially when multiple vehicles operate in close proximity.

Innovation Solution

A method and system that determine uncertainty in predicted flight paths by analyzing performance and weather models, generating a confidence score to quantify deviations, and providing notifications to air traffic control for improved air traffic management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If four-dimensional trajectories are generated by flight management system, then flight path prediction is provided, but uncertainty in weather conditions and performance models causes inaccuracy in trajectory-based operations

Engineering Contradiction:
Improveflight path prediction accuracyVSAvoidtrajectory-based operations reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent segments the uncertainty analysis into distinct components: performance model uncertainty, weather model uncertainty, and predicted flight path uncertainty. Each component is determined separately through Monte Carlo simulations, allowing for precise identification and management of different uncertainty sources that affect overall flight path prediction accuracy

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary uncertainty analysis by conducting Monte Carlo simulations before actual flight operations. By determining uncertainty in performance and weather models in advance, the system prepares confidence scores that improve the reliability of trajectory-based operations before vehicles actually execute their flight paths

Inventive Principle:
Principle #10Preliminary action

2Productivity

If multiple aerial vehicles operate in close proximity using trajectory-based operations, then airspace efficiency is improved, but uncertainty in predicted flight paths increases risk of conflict detection errors

Engineering Contradiction:
Improveairspace efficiencyVSAvoidconflict detection reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements feedback by using determined uncertainty values to generate confidence scores that are fed back into the trajectory-based operations system. This feedback mechanism allows air traffic management to adjust separation standards and conflict detection thresholds based on actual uncertainty levels, thereby maintaining high airspace efficiency while improving conflict detection reliability

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system changes operational parameters dynamically by adjusting confidence score thresholds and uncertainty margins based on determined performance and weather model uncertainties. This allows the system to optimize the balance between airspace efficiency and conflict detection reliability by modifying separation requirements and alert thresholds according to real-time uncertainty assessments

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3407332B1System and method for determining uncertainty in a predicted flight path for an aerial vehicle
Publication Date: 2023.07.12 GE AVIATION SYSTEMS LLC
  • EP3407332B1 patent drawingFigure 1
  • EP3407332B1 patent drawingFigure 2
  • EP3407332B1 patent drawingFigure 3

AI summary

A method 700 for determining uncertainty in a predicted flight path 400 for an aerial vehicle 100 can include determining 702, by one or more computing devices 602, uncertainty in a performance model of the aerial vehicle 100. The method 700 can further include determining 704, by one or more computing devices 602, uncertainty in a weather model indicative of weather conditions along the predicted flight path 400. In addition, the method 700 can include determining 706, by the one or more computing devices 602, uncertainty in the predicted flight path 400 based on the uncertainty in the performance model and the uncertainty in the weather model. The method 700 can further include generating 708, by one or more computing devices, a notification indicating the uncertainty in the predicted flight path 400.