Using consecutive point clouds for pose and motion estimation of tumbling non-cooperative target

This paper proposes a consecutive point clouds-based estimation scheme to resolve the state estimation problem for tumbling non-cooperative space target during the rendezvous phase without a prior knowledge about its structure. First, a consistent pose estimation algorithm is realized by maintaining a global structure of the target that is reconstructed upon the pose graph optimization. Then an extend Kalman filter on Lie group is adopted to estimate the motion and inertia parameters of the target using the pose measurements of the point clouds. Finally, a semi-physical experimental study is carried out to evaluate the performance of the proposed estimation scheme. The result shows that the structure, motion and the inertia parameters can be estimated, and the total computation time is approximately linear with the number of point clouds.