中文摘要：

提出了一种深空探测器着陆阶段的自主导航方法。该方法通过跟踪导航序列图像中特征点在像平面上的位置，结合激光测距仪给出的探测器到小天体表面的距离信息，在特征点实际位置不确定的情况下，利用扩展卡尔曼滤波对探测器的相对位置姿态及运动信息进行递归估计。同时重构特征点的三维位置。由于不需要所选特征点在小天体表面的精确位置，因此提高了导航方法的适用性。最后通过数学仿真验证了自主导航方法的有效性。

英文摘要：

This paper proposes an autonomous navigation method for soft landing on small body. The formulation for recursive estimarion algorithm of relative pose and motion from feature points tracked through an image sequence is presented. Due to the nonlinear of measurement equation based on the projection model of image formation, the extended Kalman filter is employed. In addition to adding distance between probe and surface of small body to state vector, the 3-D position of feature point is also estimated. The results on synthetic image show that the proposed algorithm can provide the estimation of state with satisfactory accuracy.