Determining relative position between satellites in formation is one of the most important thing to accomplish the formation flying mission. The purpose of this paper is to estimate that accurately in real-time. For this purpose, this study used GPS s...
Determining relative position between satellites in formation is one of the most important thing to accomplish the formation flying mission. The purpose of this paper is to estimate that accurately in real-time. For this purpose, this study used GPS signal, carrier-phase data, to make measurement model. Extended Kalman Filter(EKF) and Unscented Kalman Filter(UKF) are used to achieve relative navigation in real-time. Relative navigation is to estimate state vector between satellites in a formation. Dynamic model in this paper is consisted of two-body problem, J2 perturbation and air-drag. And Single differenced GPS carrier-phase data is used for measurement model. Common errors on formation satellite, chief and deputy, are eliminated by using single difference method. And this advantage of single difference can acquire more accurate measurement data. From the result in this study, relative navigation algorithm can estimate the state vector between satellites in formation under a meter level. And from this results, relative navigation system functions properly. In addition, the results from each filter, this paper compares the performance of two filters. In this study, two filters show very similar performance, but extended kalman filter is more sensitive about initial error or initial covariance matrix than uscented kalman filter. In general, RMS(Root Mean Square) data is more stable in unscented kalman filter. And using two GPS measurement data can estimate more accurate and stable state vector between satellites.