一种Kalman平滑器新算法

A NEW APPROACH OF DESIGNING KALMAN SMOOTHER

应用现代时间序列分析方法,基于ARMA新息模型和白噪声估值器,本文提出了一种Kalman平滑器新算法。它不用解Riccati方程,并且可以统一的处理预报、滤波和平滑问题,估值器具有渐近稳定性。仿真例子说明了本算法的有效性。

Using modem time- series analysis method, based on the autoregressive moving average (AR-MA)innovation model and white noise estimators, a new algorithm of Kalman smoother is presented. The computation of the Riccati equation is avoided. They can handle the prediction, filtering and smoothing problems in a unified framework, and have the asymptotic stability. Simulation example shows the validity of the new algorithms.