Novel Cubature Predictive Filter for Relative Position and Attitude Estimation of Satellite Formation Considering J2

A novel cubature predictive filter (CPF) is presented for relative position and attitude estimation of satellite formation taking into account the J2 perturbation. A coupled relative translational dynamics is derived to represent orbital motion of arbitrary feature points on deputy, and the relative attitude motion is formulated by the rotational dynamics for the satellite. Based on the proposed coupled dynamics, the CPF is developed based on the third-degree spherical-radial cubature rule, extending the study of a traditional predictive filter (PF). The algorithm flow of the CPF is proposed first. Then, it is demonstrated that the estimated accuracy of the model error and system state for CPF is higher than that of traditional PF. In addition, a cubature Kalman filter (CKF) is also employed in order to evaluate the performance of the proposed CPF. Several different scenarios are simulated to validate the effectiveness of the coupled dynamics model and the performance of the proposed CPF. Through comparisons, the proposed CPF is shown to yield to obtain more accurate relative position and attitude estimation for satellite formation.