A method of space object recognition based on threshold estimating

The aim of space object recognition is to judge whether an observed object or a new one exist or not. In conventional methods, an initial orbit is often calculated with the observed data, and then is compared with the orbit elements of existing object. This paper presents a new algorithm based on threshold estimating. The orbit of an object to be recognized is compared with the pre-calculated observable track of the existing object, and the range of identification can be reduced fundamentally. Then the aim can be achieved by making use of recognition threshold in a small range. The result of computation indicates that this method can improve the efficiency of recognition, satisfy the real-time request, and increase the dependability of recognition.