Precision pointing of imaging spacecraft using gyro-based attitude reference with horizon sensor updates

Abstract

Remote sensing satellites are required to meet stringent pointing and drift rate requirements for imaging operations. For achieving these pointing and stability requirements, continuous and accurate three-axis attitude information is required. Inertial sensors like gyros provide continuous attitude information with better short-term stability and less random errors. However, gyro measurements are affected by drifts. Hence over time, attitudes based on the gyro reference slowly diverge from the true attitudes. On the other hand, line-of-sight (LOS) sensors like horizon sensors provide attitude information with long-term stability. Their measurements however are affected by the presence of random instrumental errors and other systematic errors. The limitations of inertial and line-of-sight sensors are mutually exclusive. Hence, by optimal fusion of attitude information from both these sensors, it is possible to retain the advantages and overcome the limitations of both, thereby providing the precise attitude information required for control. This paper describes an improved earth-pointing scheme by fusion of the three-axis attitude information from gyros and horizon sensor roll and pitch measurements along with yaw updates from the digital sun sensor. A Kalman Filter is used to estimate the three-axis attitude by online estimation and corrections of various errors from the sensor measurements. Variations in orbit rate components are also accounted for using spacecraft position and velocity measurements from the satellite positioning system. Thus precise earth-pointing is achieved.