带未知观测丢失率的自校正加权观测融合估计  

Self-tuning weighted measurement fusion estimation with unknown missing measurement rate

作　　者：史腾飞 段广全 孙书利[1] SHI Teng-Fei DUAN Guang-Quan SUN Shu-Li(School of Electronic Engineering, HeilongiiangUniversity, Harbin 150080, China)

机构地区：[1]黑龙江大学电子工程学院,哈尔滨150080

出　　处：《黑龙江大学工程学报》2017年第3期71-75,共5页Journal of Engineering of Heilongjiang University

基　　金：国家自然科学基金资助项目(61573132)

摘　　要：对于带未知丢失观测率的离散线性随机系统,应用伯努利随机变量来描述观测丢失现象。采用相关函数法辨识丢失观测率。应用加权最小二乘法(WLS)把高维的观测向量进行压缩得到加权观测融合方程。将实时辨识的观测丢失率代入最优加权观测融合滤波器中得到自校正加权观测融合滤波算法。所获得的自校正加权观测融合滤波器收敛于最优融合滤波器。仿真例子验证了算法的有效性。For discrete-time linear stochastic systems with unknown missing measurement rates, the Bernoulli random variables are used to describe the phenomena of missing measurement and the correlation functions are used to identify the missing measurement rates. The weighted least squares （WLS） method is used to compress the high- dimensional measurement vector to obtain weighted measurement fusion equation. A self-tuning weighted measurement fusion filtering algorithm is obtained by substituting the real-time identified missing measurement rates into the optimal weighted measurement fusion filter. Moreover, the proposed self-tuning weighted measurement fusion filter converges to the optimal fusion filter. A simulation example verifies the effectiveness of the proposed algorithm.

关 键 词：丢失观测率 自校正 KALMAN滤波器 加权观测融合 

分 类 号：TP274.2[自动化与计算机技术—检测技术与自动化装置]