Performance analysis of the attitude-correlated frames approach for star sensors

Abstract

The attitude accuracy of star sensors decreases rapidly when star images become motion-blurred under dynamic conditions. To improve the performance of star sensors, the attitude-correlated frames (ACF) approach concentrating on the features of the attitude transforms of the adjacent star image frames, was proposed recently. It is effective in removing random noises and improving the attitude accuracy of the star sensor under different dynamic conditions and noise gray levels. In this paper, a further research on the performance analysis of the ACF approach for star sensors is presented to show its advantages and clarify its application limitations as well. With the ACF method, a much larger star image frame is obtained through the combination of adjacent frames. By increasing the number of frames N, much more information of star frames can be applied to estimate the attitude, the ACF approach therefore can output the attitude information continuously when the current techniques fail to work under highly dynamic conditions. However, the increase of computational consumption is small and acceptable for the star sensor. Since the distribution features of noises will change during the process and the error of the strap down gyro unit increases with the correlation time, an accuracy limitation will occur. The analyses are validated with simulations and experimental data. As a result, the attitude accuracy and robustness of a star sensor increase under highly dynamic conditions. These performance improvements can allow the use of star sensors during maneuvering and other dynamic conditions for satellites, spacecraft, ballistic missiles, and ships.