Star Spot Extraction for Multi-FOV Star Sensors Under Extremely High Dynamic Conditions

Abstract

On some special occasions, the spacecraft may need to change attitude at a rate of 30°/s in order to achieve a high level of mobility. To ensure star sensors can stably extract a consistent number of star spots under extremely high dynamic conditions, this article presents a method to solve the problem of accurate detection of faint star spots by low-level information fusion based on a multi-field-of-view (multi-FOV) star sensor. First, reduce the gray threshold to a low level and use the optimal directional connected component (ODCC) algorithm to extract the light spots containing star spots and false star noise. Next, establish a 3-D parameter space for Hough voting and obtain a three-axis rotation angle estimated by combining the structure characteristics and the motion characteristics with the star vector information. Utilizing the joint observation frames, the relative information of spot motion is obtained to preliminarily screen out the false stars and verify convergence during each iteration. Finally, the faint star spots can be extracted, and false star noise can be screened by the estimated three-axis rotation angle. The ground experiments have shown that compared with the traditional algorithms; our algorithm has better faint star spot extraction ability under extremely high dynamic conditions. Moreover, multi-FOV star sensors demonstrate a more robust capability than traditional single-FOV star sensors.