Javad Salem; hossein pilaram; seyyed mohammad alavi
Volume 9, Issue 4 , April 2017, , Pages 51-63
Abstract
For the detection of and tracking thelow earth orbit Satellites (LEO), there are different methods such as optic, laser and radar tracking, among which radar tracking is the best. Since the common linear tracking filters deployed in available radars are not able to estimate the position of the non-linear ...
Read More
For the detection of and tracking thelow earth orbit Satellites (LEO), there are different methods such as optic, laser and radar tracking, among which radar tracking is the best. Since the common linear tracking filters deployed in available radars are not able to estimate the position of the non-linear dynamic satellites, it is advisable to use non-linear filters. In this paper, firstly, the satellite motion path around the earth as well as radar observations are produced by the STK software. Accordingly, the samples are fed to non-linear Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF). Finally, the performance of the aforementioned filters is studied through evaluation of RMS position and estimation errors. Simulation results demonstrate that the Unscented Kalman filter has a better performance in terms of accuracy with respect to the Extended Kalman filter. In addition, using this method, theerror of observations decreases 50% along the range and 70% along the azimuth and elevation.
MohammadAli Amiri Atashgah; Hamid Gazerpour; seyed Amirreza Roghangir
Volume 9, Issue 3 , December 2016, , Pages 13-26
Abstract
This paper is dedicated to determining the orbit of a satellite by using Unscented Kalman Filter (UKF), in which, a GNSS is used as the observation sensor. During this goal, firstly we have simulated the satellite orbit; considering oblations effect. Secondly, exploiting a nonlinear model of orbit dynamics, ...
Read More
This paper is dedicated to determining the orbit of a satellite by using Unscented Kalman Filter (UKF), in which, a GNSS is used as the observation sensor. During this goal, firstly we have simulated the satellite orbit; considering oblations effect. Secondly, exploiting a nonlinear model of orbit dynamics, preliminary orbit determination is prepared via UKF algorithm. The range between the satellites and space vehicle as the parameter of observation in the filter is obtained. Afterwards, the estimated preliminary output data is corrected and more precise position of the GNSS satellite is determined based on predicted observations errors. Consequently, the outcomes of the research exhibits the acceptable satellite orbit determination error range.
