论文部分内容阅读
针对已有基于二维影像的非合作航天器定位定姿方法在目标适应性和稳定性方面的不足,研究了一种基于三维点云的位姿测量方法。提出一种考虑累积误差的参数传递法进行序列点云的配准,用参数传递法获得初始值,在初始值的引导下与基准模型进行快速精确配准,实现了基于实时三维点云的相对运动跟踪,获得了较高的解算精度,方法利用目标的几何形状信息,有更好的普适性和稳健性。分别以仿真模型和现场模型作为测量对象进行位姿跟踪实验,用扫描仪获取动态卫星模型的三维点云并进行位姿解算,测量的运动轨迹与设计值吻合度较好,表明所提方法有较好的稳健性和较高的精度。
Aiming at the deficiency of target adaptability and stability of the existing non-cooperative spacecraft positioning and orientation method based on two-dimensional images, a pose measurement method based on three-dimensional point cloud is studied. A parameter transfer method considering cumulative error is used to register the point cloud. The parameter transfer method is used to obtain the initial value, and the initial value is used for fast and accurate registration with the reference model. The relative point cloud based on real-time 3D The motion tracking achieves a higher solution accuracy. The method uses the geometric information of the target and has better universality and robustness. Respectively, the simulation model and the scene model as the measurement object to perform the pose tracking experiment. The 3D point cloud of the dynamic satellite model is acquired by the scanner and the pose solution is obtained. The measured trajectory is in good agreement with the design value, indicating that the proposed method Better robustness and higher accuracy.