David Finkleman, (Center for Space Standards and Innovation)
Keywords: Data Fusion, Conjunction Assessment, Orbit Determination
Abstract:
Different orbit determination and propagation approaches give different answers with the same data. This diversity is amplified in conjunction assessment, which is dominated by numerous random processes and uncertainties in the states of both satellites. The objective of this paper is to expose approaches to dealing with diverse, potentially equally credible predicted outcomes for the same events. One should not expect outcomes predicted by different analysis techniques to be identical even for the same events founded on the same data. Different, even controlled, realizations of the same events will almost never all have identical outcomes. Vallado has documented differences to be expected with different force models, model atmospheres, measurement uncertainties, numerical implementations, and other user specific decisions. Geopotential, Earth Orientation, and atmospheric models evolve, and outcomes with current versions will differ from outcomes using earlier instances. Alfano and Finkleman have explored causes of conjunction false alarms and missed events and bounded analysis parameters based on decision makers risk tolerance. Some analysts and warning providers will always misidentify events perceived otherwise by different providers. Who is right? This is a very common circumstance, but how it should be approached depends on the analysis goals. If the goal is the convolution of many steps each of which can be diverse among analysts, we can only strive to understand what others did to arrive at different results, try to transform all of the different predictions to as near a common set of parameters, reference frames, and other user preferences, and collaborate with others who might be involved to achieve a commonly understood course of action. No one is necessarily right or wrong. We will explore different approaches to digesting diverse predictions in order to attain common understanding and action. Which one chooses depends on the goal: pointing sensors, downlinking data or uplinking commands, accessing areas of different sizes and locations on the Earth, launch collision avoidance, conjunction assessment, and space situational awareness, at least. In many cases the diversity among predictions does not affect the use of the information and any of the diverse predictions is sufficient for the purpose intended.
Date of Conference: September 10-13, 2013
Track: Astrodynamics