Autonomous relative navigation error compensation method for small body exploration based on sequence-images

Abstract

<p indent="0mm">Deep-space small bodies are typical space non-cooperative targets. The autonomous relative navigation of space non-cooperative targets based on sequence-images is an important technical method to realize the small body exploration. Due to the complex on-orbit environment, it is inevitable that the camera will suffer the measurement deviations and the focal length deviation, which will severely affect the relative navigation state estimation accuracy. On the basis of making full use of the information of star sequence images in the field of view of the camera, the initial value of the camera on-orbit calibration is calculated by the least square method. Combined with the measurement deviation state transition model and the starlight angular distance measurement equation between two stars, the measurement deviations are expanded into state estimation parameters and the relative navigation state estimation model based on camera on-orbit error compensation is constructed. Combined with the unscented Kalman filter algorithm, the iterative estimation is carried out through comparing the information between the frames of sequence images. Then the camera measurement deviation and the relative navigation system state are obtained. In this way, the state estimation accuracy is improved. In the simulation analysis, the small body 2016HO3 is considered as the relative navigation target. The simulation results show the effectiveness of the method, which can effectively compensate the camera measurement deviation and achieve the high-precision estimation of the relative navigation system state. The research results provide certain technical support for the subsequent related tasks.