含结构参数摄动的线性离散系统的鲁棒约束方差状态估计

考虑状态矩阵中含不确定性的线性离散随机系统的鲁棒约束方差状态估计问题,即设计滤波增益,使摄动系统每个状态分量的估计误差方差的稳态值不大于各自预先给定值。给出了期望鲁棒滤波增益的存在条件及其解析表达式,并以数值算例说明了设计方法的直接性与有效性。

Robust Variance-constrained State Estimation for Linear Discrete-time Systems with Structured Parameter Perturbations

The problem of robust variance - constrained state estimation is studied for linear discrete - time stochastic systems with uncertainty in the state matrix. The aim is to design a filtering gain such that the steady -state estimation error variance for each state of the perturbed system is not more than the respective prespecified value. Existence conditions and analytical expression of the desired performance robust filtering gain are obtained. A numerical example demonstrates the directness and effectiveness of the present design approach.