Star Tracker Algorithm for Large-Angle Attitude Change

抄録

In this paper, we propose an algorithm for a star tracker (STT) for rapid, large-angle attitude maneuvers. Recently, missions with rapid large-angle changes using small satellites have been proposed. Such missions require rapid and precise changes in attitude to the required observation direction. Generally, an STT is used for high-precision attitude determination for satellites. However, STTs have angular velocity limitations and cannot determine attitude during rapid angle changes. To estimate angular velocity and attitude during attitude changes, we propose a method to identify stars using the estimated star position and previously estimated angular velocity. Furthermore, random sample consensus (RANSAC) is used for more accurate identification. Results of numerical simulation show that we can identify stars with a probability of 98.2% and estimate angular velocities with an accuracy of about 2.00 x10^-2 [deg/s]. The attitude during an attitude change can be estimated with an error angle of 1.75 x10^-2 [deg] in each axis if there is no rotation in the roll direction.