UAS Collision Avoidance via Vehicle Movement Model Trajectory Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current collision avoidance systems for Uninhabited Air Systems (UAS) face challenges in detecting aircraft in real-time with low false positives and achieving equivalent safety to human pilots, as existing methods rely on high contrast assumptions and are computationally intensive, and lack robustness in detecting non-cooperative traffic and varying angles and distances.
Innovation Solution
A method that analyzes image data to identify potential moving vehicles by comparing them with a vehicle movement model, using directional data and pixel density changes to determine trajectory and velocity, and predicts future trajectory for collision avoidance, employing a constant velocity model and 2D Gaussian weighting for accurate detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high resolution cameras are used to detect aircraft that initially appear at low resolution, then detection precision is improved, but computational complexity increases
Solution Approach 1:
The system performs preliminary actions by establishing a vehicle movement model before detailed detection. It uses directional data and pixel density changes to determine trajectory and velocity in advance, which simplifies subsequent high-resolution analysis by pre-filtering and pre-characterizing potential targets.
Solution Approach 2:
The detection process is segmented into distinct stages: initial identification using pixel density changes, trajectory estimation using movement models, and detailed analysis. This segmentation allows each stage to operate at appropriate computational levels, avoiding unnecessary complexity in early detection phases.
2Speed
If real-time detection is implemented, then response speed is improved, but measurement precision deteriorates
Solution Approach 1:
The system performs preliminary trajectory modeling and velocity estimation using constant velocity models and 2D Gaussian weighting before final detection confirmation. This preliminary action enables real-time response while maintaining precision by pre-processing motion characteristics.
Solution Approach 2:
The patent replaces traditional mechanical image processing methods with mathematical modeling approaches. It uses vehicle movement models, pixel density analysis, and Gaussian weighting functions to substitute complex mechanical detection systems, achieving both real-time performance and high precision through computational efficiency.
3Measurement precision
If GPS and scene geometry information are used, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system extracts only the essential features needed for detection - pixel density changes and directional data - while discarding unnecessary information. It takes out the core trajectory and velocity parameters from complex GPS and scene geometry data, maintaining precision through selective feature extraction rather than processing all available information.
Solution Approach 2:
The detection system uses self-service by relying on intrinsic image data properties (pixel density, directional information) rather than external GPS or scene geometry systems. The vehicle movement model is self-contained, using only the data it captures directly, which reduces system complexity while maintaining detection accuracy.
Data Source
AI summary
A method of detecting at least one moving vehicle includes receiving (202) image data representing a sequence of image frames over time. The method further includes analyzing (204-206) the image data to identify potential moving vehicles, and comparing (208-212) at least one said potential moving vehicle with a vehicle movement model that defines a trajectory of a potential moving vehicle to determine whether the at least one potential moving vehicle conforms with the model.