Satellite autonomous navigation based on magnetic field measurements

Abstract

This paper presents a near-Earth (less than 1000 km altitude) satellite autonomous navigation and orbit determination method using measurements of the Earth magnetic field. An orbit state vector comprised of six Keplerian elements enables the estimation of the instantaneous orbital elements by a relatively simple extended Kalman filter algorithm. The satellite position and velocity are computed as a function of the estimated orbital elements. Several algorithms were developed. The basic algorithm uses a measurement of the magnetic field magnitude. Consequently, this algorithm is independent of attitude information. Simulation tests yielded accurate Keplerian element estimation and a few kilometers of position estimation error. More complicated algorithms that estimate drag and/or utilize attitude information were tested. The basic algorithm was successfully applied to real Earth Radiation Budget Satellite data, and a modified version of this algorithm was applied to Gamma Ray Observatory magnetometer readings to yield estimates of the orbital elements and the position and velocity. In both cases of real satellite navigation, the position was estimated within a few tens of kilometers.