Iterative Smoother-Based Variance Estimation
Iterative Smoother-Based Variance Estimation
复制标题
基于迭代平滑器的方差估计
DOI：
--
复制
发表时间：
2012
影响因子：
3.9
通讯作者：
D. C. Reid
中科院分区：
文献类型：
--
作者：
G. Einicke;Gianluca Falco;M. Dunn;D. C. Reid
给出了输入估计的最小方差光滑解，并证明了所得估计是无偏的。平滑的输入和状态估计被用来迭代地识别未知的过程噪声方差。使用平滑估计，而不是滤波估计，改进了未知参数的近似Cramér-Rao下界。文中还证明了迭代序列是单调的，并且在一定条件下渐近逼近实际值。描述了一种估计未知参数的非线性采矿导航应用。
The minimum-variance smoother solution for input estimation is described and it is shown that the resulting estimates are unbiased. The smoothed input and state estimates are used to iteratively identify unknown process noise variances. The use of smoothed estimates, as opposed to filtered estimates, leads to improved approximate Cramér-Rao lower bounds for the unknown parameters. It is also shown that the sequence of iterates are monotonic and asymptotically approach the actual values under prescribed conditions. A nonlinear mining navigation application is described in which unknown parameters are estimated.