Abstract: Considering the relative pose estimation problem of a microsatellite autonomous approaching to an unknown and non cooperative target, a method for monocular based pose estimation of non cooperative space targets using extended Kalman filter (EKF) and extended Kalman particle filter (EKPF) is proposed. Taking the camera coordinate system of the mission satellite as reference, the relative motion model of non cooperative target is established. The measurement equation is constructed by taking the plane coordinates of feature points as the observed values. After initializing the relative pose and the three dimensional coordinates of feature points, the extended Kalman particle filter is used to estimate the relative pose, and the extended Kalman filter is used to estimate the feature locations. The non cooperative target image generated by OpenGL is used for simulation. The results show that the non cooperative pose estimation method proposed in this paper can obtain high precision pose estimation and has good robustness. Compared with the existing methods, the proposed method does not require the initial pose assumption, and can effectively reduce the computational time and improve the efficiency of the pose estimation.

Key words: Non cooperative targets, Monocular vision, Relative pose, Extended Kalman filter, Extended Kalman particle filter