Federated hybrid Extended Kalman Filter design for multiple satellites formation flying in LEO

First of all linear as well as non-linear relative dynamic model for satellites in formation flying in low earth orbit (LEO) is derived and next state estimation based on Kalman filter is emphasized. In this paper, we have proposed a more accurate and robust state estimation technique named federated extended Kalman filter (FEKF) for relative navigation information of two small satellites in formation flying scenario. Principally, state estimation in formation flying is a multi-sensor data fusion problem, so main topic of this paper is to design an optimal/suboptimal fault tolerant state estimator for this relative dynamic model which represents formation flying of two satellites in LEO. A comparison of the centralized Kalman filter (CKF) and the federated Kalman filter has been made to show that FKF exhibits fault tolerance ability. Also, FKF if used in optimal mode is equivalent to the centralized Kalman filter. Fault detection and isolation (FDI) techniques are also briefly discussed and applied to the federated filtering schemes.