Satellite Constellation RSO Detection via Contrast-Enhanced Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing number of satellites in orbit poses a risk of collisions due to space debris, which requires effective detection, identification, and mapping of resident space objects (RSOs) for collision assessment and space situational awareness.
Innovation Solution
A computer-based system and method that processes image data from a constellation of satellites to generate and map celestial image features, including RSOs, by using pre-processing pipelines for contrast enhancement and machine learning models to identify and track RSOs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the number of satellites in orbit is increased to improve communication and navigation coverage, then the quality of service is improved, but the risk of collisions with space debris increases
Solution Approach 1:
The system performs preliminary detection and tracking of space debris objects using multiple imaging satellites before collisions can occur. By continuously monitoring and maintaining an updated catalog of RSO positions and trajectories, the system enables advance warning and collision avoidance maneuvers, thus resolving the contradiction between increased satellite deployment and collision risk
Solution Approach 2:
The patent introduces an intermediary detection and monitoring system consisting of imaging satellites, image processing pipelines, and RSO catalog databases. This intermediary system acts as a mediator between the increasing number of operational satellites and the harmful collision risk, providing the information needed to manage and mitigate collision dangers
2Measurement precision
If image processing complexity is increased to improve RSO detection accuracy, then measurement precision is improved, but processing time and computational resources increase
Solution Approach 1:
The image processing pipeline is segmented into distinct modular stages: pre-processing (contrast enhancement, noise reduction), feature extraction (streak detection, object identification), and catalog update operations. This segmentation allows each stage to be optimized independently and enables parallel processing, thereby maintaining high detection accuracy while reducing overall processing time
Solution Approach 2:
The system performs preliminary image pre-processing operations including contrast enhancement and noise reduction before main analysis. Additionally, the RSO catalog is pre-populated with known object data, allowing for faster comparison and identification during real-time processing, thus improving accuracy without proportionally increasing processing time
Data Source
AI summary
A system includes a plurality of satellites orbiting around a celestial body in a plurality of orbital planes. Each satellite includes an imaging device having a field of view (FOV) to capture an image of a sky that includes celestial image features of resident space object (RSO), stars, and/or planets. The satellite processor is configured to input the image data for each image into a pre-processing software pipeline to generate for each image a contrast-enhanced image data replica of each image, and an enhanced image data replica of each image, to receive from an output of a known-unknown RSO split data processing pipeline, a determination that a candidate RSO is a known RSO stored in an RSO catalog, or an unknown RSO, and to assign to the candidate RSO based on the determination, an RSO ID of the known RSO listed in the RSO catalog, or a new RSO ID.