范数有界的鲁棒卫星姿态估计方法  

Research on satellite attitude determination based on norm bounded robust approach

作　　者：赵琳[1] 邱海洋[1] 郝勇[1] 

机构地区：[1]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《华中科技大学学报（自然科学版）》2016年第4期43-47,共5页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(B61273081);中央高校基本科研业务费专项资金资助项目(P043515002)

摘　　要：针对现有卫星姿态鲁棒滤波器中计算量大、时变系统估计效果不理想的问题,将范数有界鲁棒滤波器应用于卫星姿态估计系统中.该方法将一部分系统模型误差视为系统的不确定度,降低了系统的建模难度.通过欧几里德范数确定状态的不确定度上界,对系统进行误差有界的鲁棒估计,保证了估计精度.仿真结果表明:在假设的不准确条件下,该鲁棒滤波器展现了良好的鲁棒性,其估计性能比传统扩展卡尔曼滤波(EKF)方法在三轴的估计精度(RMS)方面分别提高了11.21%,17.97%和38.13%.For the purpose to address the computation load and undesirable performance for time-varying system in traditional robust filtering for satellite attitude estimation,a norm bounded robust filter was designed and implemented in this system.This approach treated a part of model error as system uncertainties,and hence reduced the difficulty of system modeling problem.The state uncertainties upper bound could be determined through Euclidean norm,and the error bounded robust estimation could guarantee the estimation accuracy under system uncertain case.Simulation results show that the robust norm bounded approach has an improvement of 11.21%,17.97%,38.13%respectively in terms of residual mean square（RMS）for three axis compared to extended Kalman filter（EKF）in uncertain system.

关 键 词：鲁棒滤波估计 卫星姿态确定 范数有界 黎卡迪方程 卡尔曼滤波 

分 类 号：U666.12[交通运输工程—船舶及航道工程]