| Autor: |
David Finkleman, Salvatore Alfano |
| Rok vydání: |
2014 |
| Předmět: |
|
| Zdroj: |
Acta Astronautica. 99:193-200 |
| ISSN: |
0094-5765 |
| DOI: |
10.1016/j.actaastro.2014.02.004 |
| Popis: |
This paper extends concepts of signal detection theory to predict the performance of conjunction screening techniques and guiding the selection of keepout and screening thresholds. The most efficient way to identify satellites likely to collide is to employ filters to identify orbiting pairs that should not come close enough over a prescribed time period to be considered hazardous. Such pairings can then be eliminated from further computation to accelerate overall processing time. Approximations inherent in filtering techniques include screening using only unperturbed Newtonian two body astrodynamics and uncertainties in orbit elements. Therefore, every filtering process is vulnerable to including objects that are not threats and excluding some that are threats, Type I and Type II errors. The approach in this paper guides selection of the best operating point for the filters suited to a user׳s tolerance for false alarms and unwarned threats. We demonstrate the approach using three archetypal filters with an initial three-day span, select filter parameters based on performance, and then test those parameters using eight historical snapshots of the space catalog. This work provides a mechanism for selecting filter parameters but the choices depend on the circumstances. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect