Developing A Bayesian Model Comparison Method For Analysis Of Force Models
- Paper number
IAC-07-C1.I.08
- Author
Dr. Laura Mullin, ZARM - University of Bremen, Germany
- Year
2007
- Abstract
In determining a spacecraft's orbit, a weighted least squares estimator can be used to obtain the best estimate of the trajectory from a given set of tracking data. A number of force and measurement models are required, introducing additional unknown parameters, which are incorporated into the estimation problem and are treated in a unified way with the orbital parameters. Bayesian least squares estimators have been used in some applications in order to take advantage of the potential to utilize prior information as provided by this approach. A further advantage, however, is that through the evaluation of the global likelihood for a given specification of force and measurement models, a quantitative model comparison can be made between competing model configurations. This is enabling a detailed analysis of all known forces acting on a deep-space spacecraft to be performed in a rigorous way, in an attempt to improve understanding of the spacecraft dynamics over a long-term mission.
- Abstract document
- Manuscript document
IAC-07-C1.I.08.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.