Optimal orbit determination requires physics-based process noise matrices to be taken into account when performing covariance propagation. For gravity errors of omission (potential field truncation) and commission (uncertainty of potential field coefficients), a first sequential filter algorithm, capable of making use of gravity error covariance approximations was presented by Wright et al. about a decade ago. The consideration of orbit gravity error covariance estimates has proven advantageous for long-term filter stability. However, characteristic deficiencies can be identified in the Wright et al. algorithm, which limit its performance and robustness. In this Paper, we analyze these shortcomings and their origins and show how they can be remedied. The outcome is an improved orbit gravity error covariance algorithm, which produces estimates that can directly be used with orbit determination algorithms capable of incorporating process noise. Numerical validation results, including techniques for a proper validation of errors of omission, are presented. It is shown that the developed improvements are able to correct the shortcomings of the previous algorithm.
Improved Orbit Gravity Error Covariance for Orbit Determination
Journal of Guidance, Control, and Dynamics ; 43 , 9 ; 1671-1687
2020-09-01
Article (Journal)
Electronic Resource
English
Orbit Gravity Error Covariance (AAS 08-157)
British Library Conference Proceedings | 2008
|Numerical Computation of Orbit Error Covariance Functions
Online Contents | 1996
|GEO ORBIT DETERMINATION VIA COVARIANCE ANALYSIS WITH A KNOWN CLOCK ERROR
Online Contents | 2001
|