Attitude estimation for resident space objects from light curves using spin parameter unscented Kalman filter

Date

2026-7-28

Author

Tuzcu, Gülce

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

46
views

20
downloads

Cite This

Knowledge of the attitude and spin state of resident space objects (RSOs) is important for space situational awareness, as it improves orbit prediction and supports active debris removal. Most of these objects are unresolved and appear as a single point of light. Their brightness variation over time, known as a light curve, is one of the few available sources of attitude information. Recovering attitude from a light curve is a difficult inverse problem, because the measurement is a single brightness value at each instant while the relationship between attitude and brightness is non-unique and strongly nonlinear. In this thesis, the attitude of RSOs is estimated from light curves using an unscented Kalman filter formulated with spin parameters, which are composed of the inertial spin-axis unit vector and the spin phase angle. Unlike quaternion-based filters, states in this representation vary slowly between observations. The angular velocity is included as an additional measurement alongside the apparent magnitude. Two forward models, in Blender and MATLAB, are developed using a Cook-Torrance reflection model and verified against each other. The method is applied to TOPEX/Poseidon. On Blender-generated synthetic data, the filter recovers the known spin state to within 0.015 in the spin-axis components. On three real light curves from the MMT-9 database, the estimated spin axis is consistent across tracks and the reconstructed light curves reproduce the overall shape of the observations with root mean square error values between 0.68 and 0.85 mag.

Subject Keywords

Light curve, Attitude estimation, Kalman filter, Spin parameter, Resident space objects

URI

https://hdl.handle.net/11511/120642

Collections

Graduate School of Natural and Applied Sciences, Thesis