Mass-Moment Parameter Estimation and Control of Reconfiguring Spacecraft in Attitude Testbed

Spacecraft attitude testbeds are tools for developing and validating attitude determination and control systems. A major challenge in attitude testbeds is accurately reproducing torque-free conditions by eliminating the center of mass offset. To address the challenge, this paper presents an approach on dynamic mass-moment estimation and control of varying mass properties for spacecraft attitude testbeds. The system employs a two-layer estimation framework, combining extended Kalman filters for attitude state and mass-moment parameter estimation with a Recursive Least Squares initializer. A key contribution is the ability to estimate and compensate variable mass-moment property in real-time. The spacecraft's deployable structures are simulated through an external point mass following a known trajectory with uncertainties. Numerical results demonstrate the system's performance under various operational conditions, including different levels of external mass trajectory uncertainty and system identification delays. The proposed approach offers a solution for real-time mass balancing in dynamic spacecraft configurations, with applications in ground testing of deployable structures and other variable geometries.