An information fusion algorithm for integrated autonomous orbit determination of navigation satellites

This paper focuses on the information fusion problem of integrated autonomous orbit determination using the observations from inter-satellite-link (ISL), X-ray pulsars and star sensors. A step Kalman filter structure is proposed to solve the information fusion problem of multiple subsystems that have greatly different filtering precision. The subsystems are grouped according to their measurement accuracy and the state parameters and covariance matrix of a group can be calculated using the federated filter structure and propagated to the next group step-by-step. Simulation results show that the mean user range error (URE) of the constellation will be less than 1.5 m in 60 days using the step Kalman filter structure for information fusion. And it has better performance than the federated structure in dealing with information fusion of the astronomical observations and the ISL ranging measurements in integrated autonomous orbit determination.

► The problem of information fusion using federated filter is presented and analyzed.
► A step Kalman filter algorithm is proposed to solve the information fusion problem.
► An integrated autonomous orbit determination algorithm is designed.
► Simulation results show that the step Kalman filter algorithm has better performance.