Effects of Navigation Filter Properties on Formation Flying Control
Conference Paper
Overview
Identity
Additional Document Info
Other
View All
Overview
abstract
This paper extends previous analysis on the impact of sensing noise for the navigation and control aspects of formation flying spacecraft. We analyze the use of Carrier-phase Differential GPS (CDGPS) in relative navigation filters, with a particular focus on the filter correlation coefficient. This work was motivated by previous publications which suggested that a "good" navigation filter would have a strong correlation (i.e., coefficient near -1) to reduce the semimajor axis error, and therefore, the overall fuel use. However, practical experience with CDGPS-based filters has shown this strong correlation seldom occurs (typical correlations -0.1), even when the estimation accuracies are very good. We derive an analytic estimate of the filter correlation coefficient and demonstrate that, for the process and sensor noises levels expected with CDGPS, the expected value should be very small. This observation is verified with several linear and nonlinear simulations. Further examples are presented to show that the extended Kalman filter can be forced to create solutions with stronger correlations, but these always result in larger semimajor axis errors. The combination of these simulations and analysis provide new insights on the crucial role of the process noise in determining the semimajor axis knowledge.
name of conference
AIAA Guidance, Navigation, and Control Conference and Exhibit
