Robust Adaptive Unscented Kalman Filter for Spacecraft Attitude Estimation Using Quaternion Measurements
Robust Adaptive Unscented Kalman Filter for Spacecraft Attitude Estimation Using Quaternion Measurements
复制标题
DOI：
10.1061/(asce)as.1943-5525.0000718
复制
发表时间：
2017-07-01
影响因子：
2.4
通讯作者：
Lee, Regina
中科院分区：
文献类型：
--
作者：
Lee, Daero;Vukovich, George;Lee, Regina
针对四元数测量数据在测量故障情况下的高精度姿态估计问题，提出了一种基于乘性四元数误差估计的鲁棒无迹卡尔曼滤波算法。全局姿态参数化由一个四元数给出，而局部姿态误差定义使用广义三维姿态表示。为了保证滤波器中的四元数归一化，无迹卡尔曼滤波器是由局部姿态误差推导出的乘性四元数误差方法。标准无迹卡尔曼滤波器即使对于初始大误差条件也提供足够好的估计结果。然而，在测量传感器故障的情况下，无迹卡尔曼滤波器无法提供所需的估计精度，甚至可能随着时间的推移而崩溃。该算法使用的统计功能，包括测量残差检测测量故障，然后使用一个适应计划的基础上，一个多尺度因子，使过滤器可以站在对错误的测量鲁棒性。在三种测量故障情况下，利用四元数测量值对航天器进行姿态估计。在相同的仿真条件下，将该算法与标准扩展卡尔曼滤波器和无迹卡尔曼滤波器的估计性能进行了比较。(C)2017年美国土木工程师协会。
A robust unscented Kalman filter based on a multiplicative quaternion-error approach is proposed for high-precision spacecraft attitude estimation using quaternion measurements under measurement faults. The global attitude parameterization is given by a quaternion, whereas the local attitude error is defined using a generalized three-dimensional attitude representation. To guarantee quaternion normalization in the filter, the unscented Kalman filter is formulated with a multiplicative quaternion-error approach derived from the local attitude error. A standard unscented Kalman filter provides sufficiently good estimation results even for initial large error conditions. However, in the case of measurement sensor malfunctions, the unscented Kalman filter fails in providing the required estimation accuracy and may even collapse over time. The proposed algorithm uses a statistical function including measurement residuals to detect measurement faults and then uses an adaptation scheme based on a multiple scale factor so that the filter may stand robust against faulty measurements. The proposed algorithm is demonstrated for attitude estimation of a spacecraft using quaternion measurements in three measurement fault cases. The estimation performance of the proposed algorithm is also compared with those of the standard extended Kalman filter and unscented Kalman filter under the same simulation conditions. (C) 2017 American Society of Civil Engineers.