Precise Spacecraft Attitude and Angular Velocity Estimates Using Cubature Kalman Filter on Intensely Distorted One-Axis Magnetometer Measurements

Magnetometers are widely used spacecraft attitude sensors due to their numerous advantages. Typically, fully observing a spacecraft’s attitude requires the use of at least two distinct sensor types. Thus, relying exclusively on a magnetometer introduces major challenges for estimation algorithms. The problem of spacecraft attitude estimation based on magnetometer measurements is generally nonlinear. Cubature Kalman Filter (CKF) is considered as a newly developed filter that addresses the problem of state estimation for nonlinear systems. The current research article develops a CKF algorithm that utilizes magnetometer measurements as a sole spacecraft attitude sensor. The developed algorithm provides multiple benefits over traditional methods, offering exceptional accuracy comparable to other Extended Kalman Filter based (EKF-based) algorithms. The developed CKF has a resistance to significant initial estimation errors. The proposed CKF algorithm functions in every spacecraft operational mode, consistently delivering precise results. Even when measurements are severely noisy, CKF achieves an accuracy of better than 0.24 degree approximately in each axis. This accuracy enabled the magnetometer to serve as the sole source of spacecraft attitude information despite having one or two faulty channels out of three. A benchmarking for the proposed CKF is given against many other intensely verified EKF-based algorithms to present a quantitative comparison. This comparison could help the designer of the spacecraft Attitude and Orbit Control System (AOCS) to choose an appropriate algorithm according to mission specific key performance indices. A case study spacecraft is utilized which is subject to aerodynamics drag torques, solar radiation pressure torques, and residual magnetic torques.