An Algorithm for Moving Multi-target Prediction in a Celestial Background

Abstract

Nowadays, there are thousands of in-orbit spacecrafts. It is of strategic significance to identify, track the space targets, and calculate their orbits and coverage area. In the celestial space observation of space targets on a micro spacecraft detector, it is difficult to observe and track targets because of their small size, unobvious geometric shape, and the complex movements relative to observed targets, background and detectors. The prediction algorithm of space targets must satisfy the demands of high reliability, low computation, real-time processing and so on. Based on multi-thread image processing and the Kalman algorithm, a new algorithm is proposed for multi-target prediction. Using this approach, the coordinates of targets can be predicted in real-time and precisely. Through theoretical calculation and computer simulation, this algorithm has shown that it is simple, effective and can satisfy the requirement of working conditions of the micro spacecraft detector well.