Session
Session X: Ground Systems
Location
Utah State University, Logan, UT
Abstract
Spacecraft attitude testbed has been a popular platform for the development, testing, and validation of real-time attitude determination, estimation, and control, algorithms. The focus of this paper is the development of a dynamic mass moment estimation algorithm and an automatic mass balancing control system with numerical validations. Two estimation layers are implemented: one for attitude estimation and another for mass-moment parameter estimation. Attitude estimation follows the existing literature of quaternion estimators with a nonlinear Kalman filter. Dynamic mass-moment parameter estimation is formulated as batch estimation with a nonlinear filter. The initialization process of the EKF parameter estimator is enhanced by a recursive least squares (RLS) estimator to minimize the error covariance of the initial guess. The unscented transform is used for uncertainty propagation in the measurement model. This two-layer framework enables dynamic estimation of the spacecraft’s mass-moment parameters while the mass balancing system is actively moving. The simulation results show that the integrated nonlinear estimation techniques provide a dynamic solution for automatic mass balancing, adapting to changes in mass distribution and maintaining balanced performance. The proposed methodology offers a baseline approach for future research and development in dynamic estimations of real-time mass balancing.
Estimation and Control of Dynamic Mass-Moment Parameters for Spacecraft Attitude Simulator
Utah State University, Logan, UT
Spacecraft attitude testbed has been a popular platform for the development, testing, and validation of real-time attitude determination, estimation, and control, algorithms. The focus of this paper is the development of a dynamic mass moment estimation algorithm and an automatic mass balancing control system with numerical validations. Two estimation layers are implemented: one for attitude estimation and another for mass-moment parameter estimation. Attitude estimation follows the existing literature of quaternion estimators with a nonlinear Kalman filter. Dynamic mass-moment parameter estimation is formulated as batch estimation with a nonlinear filter. The initialization process of the EKF parameter estimator is enhanced by a recursive least squares (RLS) estimator to minimize the error covariance of the initial guess. The unscented transform is used for uncertainty propagation in the measurement model. This two-layer framework enables dynamic estimation of the spacecraft’s mass-moment parameters while the mass balancing system is actively moving. The simulation results show that the integrated nonlinear estimation techniques provide a dynamic solution for automatic mass balancing, adapting to changes in mass distribution and maintaining balanced performance. The proposed methodology offers a baseline approach for future research and development in dynamic estimations of real-time mass balancing.
