处理不确定性姿态确定系统误差的新型鲁棒滤波方法  

A Novel Robust Filtering Method with Uncertainty Attitude Determination System Rrror

作　　者：矫媛媛 潘晓刚[1] 雷拥军[2] 张春青[2] JIAO Yuan-yuan;PAN Xiao-gang;WANG Jiong-qi;ZHOU Hai-yin(College of System Engineering,National University of Defense Technology,Changsha 410073,China;Beijing Institute of Control Engineering,Beijing 100080,China)

机构地区：[1]国防科技大学系统工程学院,湖南长沙410073 [2]北京控制工程研究所,北京100080

出　　处：《模糊系统与数学》2022年第1期157-164,共8页Fuzzy Systems and Mathematics

摘　　要：针对卫星姿态确定系统中存在不确定性系统误差问题,本文提出了一种适于处理不确定性系统误差的新型鲁棒滤波方法,以提高姿态确定精度。在深入分析了经典的鲁棒姿态确定方法的优势及缺陷基础上,利用相对安装误差等系统误差之间的相关性信息,将不确定性误差降维处理,进而,基于最小化估计误差的方差上界的思想,设计出改进的新型鲁棒滤波方法,抑制不确定性系统误差的影响。最后通过仿真实验验证了本文所提出方法的有效性以及相对于经典的卡尔曼滤波(EKF)方法、扩维的EKF(E-EKF)方法以及经典的鲁棒滤波方法的优越性。A novel robust filtering method was proposed to deal with the uncertainty system error for attitude determination in this paper.Based on the analysis of classical robust filter’s advantage and shortcoming,an improved new robust filtering method was designed.In this new method,relativity of the relative mounting error was used to reduce the dimension of uncertainty system error.Then the new method was designed according to minimize the upper limit of estimated error’standard deviation.Simulated experiments were used to validate the proposed method.The results showed the outperformance of this new method compared with classical Extended Kalman Filter(EKF),Extended EKF and classical robust filtering method.

关 键 词：卫星姿态确定 不确定性系统误差 鲁棒滤波 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]