针对近距离编队飞行中位姿估计问题,提出了一种新的点到区域匹配航天器间视觉相对位姿估计算法.给定主星表面上的点集及与之匹配的从星摄像机像平面图像凸区域,结合对偶四元数及凸优化数学工具,充分利用对偶四元数描述坐标系变换的简洁性,建立了对偶四元数凸优化位姿估计模型,估计出主星与从星之间的相对位姿参数.该算法不仅利用对偶四元数较传统四元数描述坐标系变换的优势,而且采用凸优化方法可大大降低传统位姿估计方法中需要精确的点到点匹配的要求.仿真结果证明,该算法能满足近距离编队相对位姿估计精度要求,具有较好的鲁棒性,对目前航天任务中航天器间视觉相对位姿估计具有重要参考价值.%A novel vision-based relative position and attitude estimation algorithm using points to regions correspondence is presented for close formation flying. Given the points on the leader satellite surface and the convex regions in which the correspondent image points lie on the follower satellite image plane, combining dual-quaternion and convex optimization mathematical tools to take full advantage of the simplicity of the dual quater-nion description of the coordinate system transformation, the convex optimization problem representation is for-mulated by using dual-quaternion representing the rotation and translation, by means of which the relative posi-tion and attitude parameters are estimated. This algorithm uses the advantage of the dual quaternion description which is better than the traditional quaternion coordinate system transformation, and using convex optimization can reduce the requirement of precise point to point correspondence in the traditional algorithm. Simulation shows that the algorithm can meet the demand of relative pose estimation accuracy and has a good robustness for close formation flying, and this algorithm can provide a reference for vision relative pose estimation between va-rious spacecraft in the current space mission.
展开▼
