Autonomous navigation algorithm for precision landing based on computer vision

摘要

In this paper we propose a visual algorithm for use by a deep space exploration spacecraft to estimate the relative position and attitude on broad during the descent phase. This algorithm is composed of the relative motion recovery which provides part motion states estimates based on tracking feature through the monocular image sequence, and landmark recognition based algorithm which offers the scale of the relative motion and absolute position of spacecraft. The results on synthetic image show that the proposed algorithm can provide the estimation of state with satisfactory accuracy.