Automatic evaluation of the internal parameters of the onboard camera of a spacecraft from video data of dockings with the iss

Abstract

The KURS radio engineering system for measuring motion parameters during rendezvous and docking has some disadvantages: the accuracy of measurement with multiple reflections of the wave can drop, the technical equipment is available on both docking vehicles (active and passive parts), it is expensive both in terms of energy resources and in terms of cost. An analysis of existing visual systems has shown that such systems successfully solve the problems of visual odometry on UAVs, robots, and similar devices. However, to use such systems, it is necessary to know the internal parameters of the camera (calibration). Classical calibration using a checkerboard pattern is difficult to perform in outer space. In connection with all of the above, this paper proposes methods for estimating the focal length of the camera, based on the analysis of the available video sequence with the footage of the process of rendezvous of spacecraft. The proposed approaches are based on the maximum likelihood method (MLE) and maximum a posteriori estimation (MAP) of the functional depending on the Euler angles and focal length. The results of these methods are compared, showing the advantages of MAP over MLE and the possibility of their practical application.