Nicholas Ravago, Omitron, Inc.; Eliot Toumey, Omitron, Inc.; Luis Baars, Omitron Inc.; Matthew Hejduk, The Aerospace Corporation
Keywords: Conjunction Assessment, Screening Volumes
Abstract:
Space object collision detection can be a computationally demanding process due to the number of conjunctions that occur between the thousands of catalogued objects in orbit. For each conjunction, a collision probability calculation must be performed to gauge whether the collision risk is high enough to warrant a remedial action. To alleviate computational load, analysts may employ a Screening Volume (SV) to filter out conjunctions in which secondary objects do not pass sufficiently close to the primary object. This SV is typically ellipsoidal in shape with semi-major axes on the order of kilometers to tens of kilometers long, and can vary with altitude due to the differing dynamics in different orbital regimes.
Choosing the appropriate screening volume size presents a tradeoff between computational load and performance. Smaller SVs filter out more conjunctions and reduce computational load, but there is a risk that a high-risk conjunction may be filtered out. In this paper, we evaluated the screening volumes currently employed by NASA’s Conjunction Assessment Risk Analysis (CARA) program. Our goal was to investigate if computational savings could be achieved via smaller screening volumes without compromising the performance levels of current screening volumes. We also wanted to develop guidelines on how SV dimensions should be adjusted for operators that want to pre-screen for a planned maneuver.
Our analysis evaluated a range of different screening volumes for their ability to correctly identify high-risk events with the objective of identifying the smallest and most efficient screening volumes that achieve a desired performance threshold. We first selected a set of primaries spanning the various orbital regimes employed by NASA CARA and used the SuperCOMBO (Computation Of Miss Between Orbits) analysis tool to identify conjunctions between these primaries and the rest of the space object catalog over the four-year period from 2020-2024. Collision probabilities were calculated for each conjunction to identify high-risk or “red” events (probability of collision greater than or equal to .0001.) Since red events generally comprise an extremely small portion of total conjunctions, risk calculations were performed using inflated Hard-Body Radius (HBR) values. This not only produced a larger set of red events, but also prevented performance from being biased by small objects which generally yield lower risk values. Since screening volume ellipsoid dimensions can be adjusted independently along all three axes, we generated an efficient frontier of possible screening volumes that achieve a desired performance threshold (e.g. 95% of all red events) for a given orbital regime. Finally, we selected the screening volume dimensions from this frontier that yielded the lowest number of total conjunctions processed.
We found that NASA could make significant reductions to their current screening volume sizes and maintain current levels of red event detection while significantly reducing the total number of conjunctions processed. The potential computational savings range from 50% in Low-Earth Orbit (LEO) regimes to over 90% in MEO/GEO regimes. In general, we found that total computations were most sensitive to the radial screening volume dimension.
We also developed guidelines for increasing SV dimensions for operators who want to pre-screen for a planned maneuver. These guidelines were based on flight dynamics principles and maneuver history data from NASA Earth Observing System (EOS) satellites. These maneuver-inflated SV dimensions can increase total conjunctions processed by up to 200% compared to current nominal SVs.
Acknowledgements
This work was funded by NASA CARA and the Office of Space Commerce (OSC) to assess the screening volume approach to be used for the TraCSS program and is related to a TraCSS technical and operation description paper being prepared by OSC for AMOS.
Date of Conference: September 16-19, 2025
Track: Conjunction/RPO