Real-Time Attitude Determination of a Nanosatellite using GPS Signal-To-Noise Ratio Observations

In support of the FASTRAC nanosatellite mission, a COTS, single antenna GPS receiver has been augmented for use in space as a multi-purpose navigation sensor. In addition to providing measurements of position and velocity, the Mitel Orion GPS receiver has been coupled with a three-axis magnetometer to provide robust attitude determination for the FASTRAC nanosatellite pair. An algorithm is presented for attitude determination of small spacecraft using single antenna GPS signal-to-noise ratio observations coupled with a magnetometer. Real-time accuracies of 5-7 degrees RMS are demonstrated in simulation. In addition, a benchmark testing procedure for evaluating the on-orbit performance of the receiver is presented. The procedure is used to characterize the raw measurement accuracy and systematic tracking loop errors for the Orion receiver. An on-orbit demonstration of the integrated sensor is planned for 2006. The integrated device is intended as a low-cost, standard solution for use on small spacecraft. Algorithm and hardware simulation results are provided to show the usefulness, accuracy, and robustness of this approach. INTRODUCTION The utilization of Global Positioning System (GPS) signal-to-noise ratio measurements (SNR) from a single antenna receiver is an innovative approach for attitude determination of LEO spacecraft. For spacecraft that are orbiting below the GPS constellation, GPS sensors have proven to be a low cost solution for orbit determination. In addition to requirements on position and velocity, most spacecraft have pointing requirements necessary for completion of their mission objectives. For instance, proper orientation of spacecraft solar arrays with respect to the sun, of cameras or other instruments with respect to objects of interest, and of directional antennas toward communication centers are all often critical for mission success. Presented here is a method for combining GPS with a three-axis magnetometer (TAM) in order to provide relatively accurate and robust attitude solutions. A least-squares extended Kalman filter (EKF) algorithm has been utilized to estimate the attitude of the spacecraft in quaternion form. After each measurement update, the spacecraft attitude quaternion is then transformed into roll, pitch, and yaw angles, and both the quaternion and Euler angles are reported for use by the attitude control system. This algorithm was initially utilized to process GPS signal-to-noise ratios (SNR) in tandem with GPS differential carrier phase measurements between multiple antennas [1]. However, the present algorithm has been modified to support a single antenna SNR solution coupled with a magnetometer for use on small spacecraft. Because this algorithm requires only a single GPS antenna, the spacecraft surface area requirements as well as the GPS hardware complexity are greatly reduced. Magnetometers alone can be used to determine the attitude of low-Earth orbiting spacecraft. However, the measurements supply only a single vector point of reference. Thus, attitude solutions can only be computed after a sufficient amount of motion through the Earth’s magnetic field has occurred. The combination with GPS signal-to-noise ratio measurements has eliminated this requirement. A combined GPS and magnetometer attitude determination setup would be attractive where space is at a premium and it is not possible to utilize an array of GPS antennas, such as on a nanosatellite. A GPS receiver is generally required for such a satellite to determine position, velocity, and time. In addition, the SNR measurements, when coupled with a magnetometer, can also provide low cost, light weight, and power efficient way to determine the attitude of the vehicle.