中文摘要：

针对探测器着陆小天体的过程提出了一种基于特征匹配结合惯导信息的探测器自主着陆导航算法。采用PCA-SIFT算法提取绕飞段拍摄小天体图像的特征点,建立特征点数据库,在下降段初期利用特征匹配算法,结合小天体三维模型确定探测器在小天体固连坐标系中的位置。着陆末端,利用拼接图像作为虚拟导航地图,并选择合适的导航特征点,采用光学/惯性的组合导航方式估计探测器相对着陆点坐标系的6自由度位姿。通过仿真分析了导航的精度及误差影响,验证了该算法的有效性。

英文摘要：

A method for spacecraft autonomous landing on the asteroid is presented in this paper bases on fusion of feature points database matching combined with inertial navigation. Feature points can be extracted from fly-around images by PCA-SIFT algorithm while the feature points database can be built. At the beginning of landing,the position and attitude of a spacecraft can be estimated by using feature points matching and 3D small body shape. At the terminal landing,takeing the mosaic image as virtual navigation map and selecting appropriate points,the 6DOF states of the spacecraft can be estimated by the combination of optical navigation and INS. And then a combined optical inertial navigation approach is given and the position of current image in the mosaic image is calculated by using the selected points matching. Finally,the navigation accuracy and error effect are analyzed and the validity of the proposed scheme is confirmed by computer simulation.