Non-Cooperative Target Pose Estimation based on Improved Iterative Closest Point Algorithm

For localisation of unknown non-cooperative targets in space, the existence interference points causes inaccuracy pose estimation while utilizing point cloud registration. To address this issue, paper proposes a new iterative closest (ICP) algorithm combined with distributed weights to intensify dependability and robustness target localisation. As space have not yet been extensively studied, we classify them into two broad categories, far near points. former, statistical outlier elimination is employed. latter, Gaussian weights, simultaneously valuing variation Euclidean distance from each centroid, are commingled traditional ICP algorithm. In iteration, weight matrix $W$ connection overall obtained, singular value decomposition adopted accomplish high-precision pose. Finally, experiments implemented by shooting satellite model setting position The outcomes suggest that proposed can effectively suppress enhance accuracy estimation. When number reaches about 700, average error angle superior 0.88°.