Interlaced optimal-REQUEST and unscented Kalman filtering for attitude determination  

作　　者：Quan Wei Xu Liang Zhang Huijuan Fang Jiancheng 

机构地区：[1]Science and Technology on Inertial Laboratory, Key Laboratory of Fundamental Science for National Defense-Novel Inertial Instrument & Navigation System Technology, Beihang University

出　　处：《Chinese Journal of Aeronautics》2013年第2期449-455,共7页中国航空学报（英文版）

基　　金：co-supported by the National Natural Science Foundation of China (Nos. 61004140, 61004129, 60825305, 61104198, 60904093);National Basic Research Program of China (No. 2009CB7240 0101C)

摘　　要：Aimed at low accuracy of attitude determination because of using low-cost components which may result in non-linearity in integrated attitude determination systems, a novel attitude determination algorithm using vector observations and gyro measurements is presented. The various features of the unscented Kalman filter （UKF） and optimal-REQUEST （quaternion estimator） algorithms are introduced for attitude determination. An interlaced filtering method is presented for the attitude determination of nano-spacecraft by setting the quaternion as the attitude representation, using the UKF and optimal-REQUEST to estimate the gyro drifts and the quaternion, respectively. The optimal-REQUEST and UKF are not isolated from each other. When the optimal-REQUEST algorithm estimates the attitude quaternion, the gyro drifts are estimated by the UKF algorithm synchronously by using the estimated attitude quaternion. Furthermore, the speed of attitude determination is improved by setting the state dimension to three. Experimental results show that the presented method has higher performance in attitude determination compared to the UKF algorithm and the traditional interlaced filtering method and can estimate the gyro drifts quickly.Aimed at low accuracy of attitude determination because of using low-cost components which may result in non-linearity in integrated attitude determination systems, a novel attitude determination algorithm using vector observations and gyro measurements is presented. The various features of the unscented Kalman filter （UKF） and optimal-REQUEST （quaternion estimator） algorithms are introduced for attitude determination. An interlaced filtering method is presented for the attitude determination of nano-spacecraft by setting the quaternion as the attitude representation, using the UKF and optimal-REQUEST to estimate the gyro drifts and the quaternion, respectively. The optimal-REQUEST and UKF are not isolated from each other. When the optimal-REQUEST algorithm estimates the attitude quaternion, the gyro drifts are estimated by the UKF algorithm synchronously by using the estimated attitude quaternion. Furthermore, the speed of attitude determination is improved by setting the state dimension to three. Experimental results show that the presented method has higher performance in attitude determination compared to the UKF algorithm and the traditional interlaced filtering method and can estimate the gyro drifts quickly.

关 键 词：Attitude determination Hybrid simulation Interlaced filtering Optimal-REQUEST Unscented Kalman filter 

分 类 号：V448.22[航空宇航科学与技术—飞行器设计]