A Switch Unscented Kalman Filter for Autonomous Navigation System of DSS Based on Relative Measurements

The autonomous navigation of distributed satellite system (DSS) using only on-board measurements a basic task for various space missions. The main purpose of this paper is to investigate new methods of autonomous navigation algorithm based on relative position measurements. A switch unscented Kalman filter (UKF) is proposed to solve the overstaffing measurement and high cost of large scale DSS. The observation switches according to the current observable degree of each subsystem witch consist of two satellites. Numerical simulations are carried out to compare the performance of switching UKF and traditional UKF. It is demonstrated that the switch UKF can greatly reduce the number of sensors and system cost on the basis of sufficient filtering accuracy.