| 摘要: |
| 针对深空探测星际着陆过程中的自主导航问题，提出一种基于星表特征直线的着陆器位姿估计算法。该算法首先采用EDLine算法对着陆过程中所拍摄的图像进行特征直线提取；其次根据直线局部特征对特征直线进行匹配；之后利用至少3对已匹配的特征直线，建立关于着陆器位姿的几何约束方程；然后根据奇异值分解法得到着陆器位姿的候选解；最终通过最小二乘法从候选解中选取着陆器位置、姿态的唯一解。仿真结果表明：该算法可以快速估计着陆器位姿，且在高度为2 000 m时位置误差小于2 m、姿态误差小于0.10°。 |
| 关键词: 特征直线匹配 视觉导航 深空探测 直线描述符 星际着陆 |
| DOI：10.15982/j.issn.2095-7717.2017.03.013 |
| 分类号:V41 |
| 基金项目: |
|
| A New Approach Based on Line Correspondences for Attitude and Position Estimation of Lander |
|
SHAO Wei, GU Tianhao
|
|
College of Automation & Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266042, China
|
| Abstract: |
| In order to achieve precise pin-point landing in deep space exploration missions，a new approach is presented to estimate accurately the lander’s attitude and position based on line correspondences during descent phase. Firstly，lines on the images taken by onboard navigation camera are detected and matched in this algorithm. The linear equations are deduced by n lines and their projection lines. Then，according to the Singular Value Decomposition，the candidate solutions are computed. Lastly，the unique solution about the attitude and position of the lander is obtained by using the least square method. The extensive experiments demonstrate that the error of attitude is less than 0.1° and the error of position is less than 2 m. |
| Key words: attitude and position estimation line correspondences visual navigation deep space exploration mission line band |
