Orbital Conjunction Prediction Using LSTM for Early Collision Warning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems are unable to accurately predict and visualize conjunctions between orbiting bodies in low Earth orbit with sufficient lead time, necessitating improvements in orbital prediction and collision warning to mitigate potential collisions.
Innovation Solution
A method utilizing a long short-term memory (LSTM) recurrent neural network (RNN) to process historical orbit data, predict future paths, and identify potential conjunctions, enabling warnings 30 or more days in advance, with remedial actions initiated through satellite management systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If state-of-the-art conjunction prediction systems are used, then prediction accuracy is improved, but prediction lead time is limited to approximately 3 days due to computing resource requirements
Solution Approach 1:
The system segments the conjunction prediction process into multiple stages: initial coarse prediction using simplified models to identify potential conjunctions, followed by refined prediction using more accurate models only for those specific cases. This segmentation allows the system to maintain high accuracy while reducing overall computational burden and extending lead time.
Solution Approach 2:
The system performs preliminary filtering and coarse prediction using computationally efficient models before applying resource-intensive accurate prediction models. This preliminary action identifies and prioritizes only those orbital paths that require detailed analysis, enabling the system to extend prediction lead time while maintaining accuracy for critical conjunctions.
2Adaptability or versatility
If the number of orbiting bodies increases to accommodate mega-constellations, then commercial connectivity capability is improved, but collision risk increases
Solution Approach 1:
The system performs preliminary identification and classification of orbital paths, predicting potential conjunctions in advance using efficient algorithms. This preliminary action enables proactive collision avoidance planning for large mega-constellations, allowing operators to coordinate maneuvers before conflicts arise, thus supporting increased satellite density while mitigating collision risk.
Solution Approach 2:
The system continuously monitors orbital paths and provides real-time feedback on conjunction risks to satellite operators. This feedback mechanism enables dynamic adjustment of orbital maneuvers and collision avoidance strategies, allowing mega-constellations to maintain safe operations despite the increased number of orbiting bodies.
Data Source
AI summary
The ever increasing number of orbiting bodies in low Earth orbit has made it infeasible to calculate potential conjunctions between orbiting bodies more than a few days in advance, even with the aid of supercomputers. Disclosed embodiments utilize machine learning to predict potential conjunctions between orbiting bodies faster than state-of-the art systems by orders of magnitude. This enables potential conjunctions to be identified well in advance (e.g., 30 days or more), so that they may be prioritized (e.g., for fine calculations), visualized, and remediated (e.g., via control of the impacted satellites).