High performance computing satellite orbit determination using ground station observations

Abstract

This article aims to execute a high performance and efficient orbit determination computation for satellites, by using least-squares algorithms as the technique of computing applied to observations from a ground station, and then the performance of the orbit estimation was analyzed. To accomplish this task, we have developed a force model encompassing these capabilities: high degree and order for the ge-potential coefficients; drag coefficient; solar radiation pressure; and Sun-Moon-planets attraction. A real state vector and observations of azimuth, elevation, and range (AER) data from the ground station were used as an input to the batch least-squares orbit determination process to offer precise results. The achieved results were compared with the orbit determination kit module (ODTK) results and the real position and velocity of the satellite, the comparison showed better precision for the adopted application's results than the ODTK results.