基于ARMA新息模型与Riccati方程的两种Kalman跟踪滤波器的等价性

针对带位置和速度观测的目标跟踪系统，分别利用ARMA新息模型法和Riccati方程法求稳态Kalman滤波器增益，通过仿真验证了两种方法的等价性。在构造ARMA新息模型时，必须进行多项式矩阵的左素分解，才能得到正确的ARMA新息模型，否则将引出错误的滤波结果。

Equivalence of Two Kinds of Kalman Tracking Filters Based on the ARMA Innovation Model and Riccati Equation

r the target tracking system with the position and velocity measurements, the steady-state Kalman filter gain is computed by two methods based on the ARMA innovation model and based on the Riccati equation, respectively. The equivalence of two methods is verified via simulation. Notice that constructing the ARMA innovation model, a left-coprime factorization to a polynomial matrix must be performed, so that the ARMA innovation model can correctly be obtained. Otherwise the mistaking filtering result is follow on.