Orbital State Estimation Based On The Extended Kalman Filter With Emergency Scenario

<p dir="ltr">The Extended Kalman Filter (EKF) based on both central body (CG) and J2 perturba-</p><p dir="ltr">tion (J2) gravity models were implemented to estimate the orbital state of a satellite in orbit</p><p dir="ltr">around Earth. The estimate was compared to the trajectory generated by General Mission</p><p dir="ltr">Analytical Tool (GMAT) which was assumed to be a real state. The onboard measure-</p><p dir="ltr">ment was provided by the Global Navigation Satellite System (GNSS) receiver. The EKF</p><p dir="ltr">performance was verified by considering unfiltered measurement signals; it was found that</p><p dir="ltr">the EKF reduces measurement error by up to 90%. To simulate an emergency scenario, the</p><p dir="ltr">measured state was temporarily disengaged, and the capability of EKF to maintain low state</p><p dir="ltr">estimation error was investigated. The simulation results proved that the state estimation</p><p dir="ltr">of CG and J2 models are comparable if the satellite has continued access to measured data.</p><p dir="ltr">The emergency study, however, found that the J2 model produces more accurate results by</p><p dir="ltr">at least one order of magnitude as compared to the CG model. To investigate how quickly</p><p dir="ltr">the true state can be restored after the emergency, the recovery time was examined and</p><p dir="ltr">found to be on the order of 10 – 20 s for all cases studied.</p>