摘要
针对“中国空间站”等复杂结构航天器在轨姿态估计误差精度优化的问题,提出了一种基于多线性结构特征优化选择的姿态估计方法。首先采用奇异值分解方法获取ISAR等效成像平面的空间位置,然后通过Unet3+神经网络分割出航天器的典型部件,进而获取各部件的线性结构,最后建立优化函数求解航天器在轨姿态角。仿真和实测实验结果表明,相比现有方法,所提方法能够实现结构特征的自动提取并有效减少姿态估计误差。
Aiming at the problem of in orbit attitude estimation error accuracy optimization of complex structure spacecraft such as‘China Space Station’,an attitude estimation method based on multi-linear structure feature optimization selection is proposed.Firstly,the spatial position of the equivalent imaging plane is obtained by singular value decomposition(SVD).Then the typical components of the spacecraft are segmented by Unet3+neural network,so the linear structure of each component is obtained.Finally,the attitude of the spacecraft in orbit is solved by an optimal estimation problem.Simulation and experimental results show that compared with the existing methods,the proposed method can realize the automatic extraction of structural features and effectively reduce the attitude estimation error.
作者
寇鹏
刘永祥
张弛
李玮杰
张双辉
霍凯
KOU Peng;LIU Yongxiang;ZHANG Chi;LI Weijie;ZHANG Shuanghui;HUO Kai(School of Electronic Science,University of Defense Science and Technology,Changsha 410073,China;Xi’an Satellite Control Center,Xi’an 710600,China)
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2023年第8期2438-2445,共8页
Systems Engineering and Electronics
基金
国家自然科学基金(61921001,61801484)资助课题。
关键词
逆合成孔径雷达
姿态估计
成像平面
部件分割
inverse synthetic aperture radar(ISAR)
attitude estimation
imaging plane
component segmentation
