Performance Tests of Two Precision Attitude Determination Systems

Abstract

Results of laboratory performance tests of two satellite attitude determination systems are given. One system employed a strapdown star tracker and gyro assembly, the other a single-axis, gimbaled star tracker and a gyro assembly. The laboratory tests simulated those orbit conditions that would be experienced on a three-axis stabilized, earth pointed satellite in geosynchronou s orbit. A ground-fixed laboratory test was performed in which system axes remained stationary in the laboratory coordinates while revolving star beams stimulated the star trackers. The laboratory instrumentation techniques used to meet the stringent accuracy requirements are described. Results are presented which show both systems met the performance goal of 3.6 arc-s. Comparative analyses of both systems are also discussed. I. Introduction T HE need for precision attitude determination systems (PADS) on board satellites has been well established by the requirements of astronomical and earth-pointing payloads over the last decade. With growing frequency PADS accuracy requirements are extending to the arc-second level. Based on early design and analysis efforts,lj2 two such PADS have been developed within the last few years: one employing a strapdown star tracker,3 and gyros, the other using a single-axis, gimbaled star tracker and gyros. The gimbaled tracker design was based on an earlier two-axis version.4 Both systems were designed for an accuracy of 3.6 arc-s (la) per axis. The objective of the effort described in this paper was a laboratory evaluation of both PADS at the system level. Previous efforts57 to demonstrate this accuracy had shown the difficulties introduced by laboratory instrumentation errors. However, a test simulating a geosynchronous orbit application presents an opportunity to minimize such errors, permitting test results to be recorded as measured, rather than being modified for instrumentation errors. The primary focus of this effort was on the error contribution of the star trackers, and the effectiveness of the software algorithms in functioning with actual sensor signals. II. PADS Description A satellite attitude determination system provides a measurement of the inertial attitude of its reference axes. The attitude of the spacecraft or payload axes is determined through a known transformation to the PADS reference axes. A functional block diagram of PADS is shown in Fig. 1. The sensors consist of 1) a star tracker that provides periodic updates of attitude relative to identified stars, and 2) gyros that provide a continuous indication of the relative inertial attitude. The sensor signals are processed in a digital computer (either on board or ground based). The computer software algorithms combine the redundant sensor signals and compute the inertial attitude of the PADS reference axes. In orbit, a strapdown star tracker uses the orbit rate of the satellite for star availability. A gimbal star tracker has the