带量测缺失的线性离散系统的滤波算法研究

研究了含有量测缺失的线性离散系统未知干扰和状态估计问题。首先将量测丢失建模为带二进制变量的伯努利过程,来模拟量测丢失的随机情形;其次考虑测量方程中未知干扰的系数矩阵不满秩和量测信息缺失的情况,设计一种抗干扰滤波器,该滤波器满足了未知干扰和系统状态估计的最小方差无偏性;最后通过数值仿真验证了在发生量测信息缺失的不同概率下,仍能实现对系统状态和未知干扰的无偏估计,表明了所提算法的有效性。

Research on filtering algorithm for linear discrete systems with missing measurements

This paper is concerned with the problem of state and unknown disturbance estimation for linear discrete systems with missing measurements. Firstly, the measurement loss is modeled as a Bernoulli process with binary variables to simulate the random case of measurement loss. Considering that the coefficient matrix of unknown disturbance in the measurement equation is not of full rank and the measurement information is missing, an anti-disturbance filter is designed, which satisfies the minimum variance unbiasedness of unknown disturbance and system state estimation. Under different probability of missing measurement information, unbiased estimation of system state and unknown disturbance can still be realized, a numerical simulation is presented to demonstrate the effectiveness of the proposed algorithm.