Angular velocity estimation using characteristics of star trails obtained by star sensor for spacecraft

Many studies have been conducted about angular velocity estimation using star sensors, and fairly good performance is achieved when the spacecraft is working at low-dynamic environments. However, when the spacecraft rotates at a large angular velocity, the star image will become blurred, which makes it difficult to identify and recognize star centroids, which, in turn, reduces the accuracy of angular velocity estimation. Therefore, to solve the problem with angular velocity estimation in highly dynamic situations, this research studies a method of angular velocity estimation using blurred star images. The length and ending point coordinates of star trails obtained from these blurred star images are used for a series of processes, which includes the pre-processing of blurred star images, determination of starting and ending points, thinning and selection of star trails, and calculation of trail lengths. Simulations show that the effectiveness of the proposed method in highly dynamic situations and the angular velocity estimation errors in constant and sinusoidal variation are reduced by at least 66% and 62%, respectively, compared with those of the traditional method.