Technical Communique: Multi-sensor information fusion white noise filter weighted by scalars based on Kalman predictor
Technical Communique: Multi-sensor information fusion white noise filter weighted by scalars based on Kalman predictor
复制标题
DOI：
10.1016/j.automatica.2004.03.012
复制
发表时间：
2004-08
期刊：
影响因子：
6.4
通讯作者：
Shu-Li Sun
中科院分区：
文献类型：
--
作者：
Shu-Li Sun
在线性最小方差意义下，提出了一种统一的标量加权多传感器最优信息融合准则。该准则考虑了局部估计误差之间的相关性，只需要计算标量权值，避免了矩阵权值的计算，从而显著降低了计算量。基于该融合准则和Kalman预报器，针对多传感器测量的离散时变线性随机控制系统，给出了输入白色噪声下的最优信息融合滤波器，可应用于石油勘探地震数据处理中。它具有两层融合结构。第一融合层具有网状并行结构，以确定在每个时间步长处任意两个传感器之间的状态的第一步预测误差互协方差和输入白色噪声的滤波误差互协方差。第二层融合层作为融合中心，确定最优标量权值，得到输入白色噪声的最优融合滤波器。对Bernoulli-Gaussian白色噪声滤波器的两个仿真例子表明了该方法的有效性。
A unified multi-sensor optimal information fusion criterion weighted by scalars is presented in the linear minimum variance sense. The criterion considers the correlation among local estimation errors, only requires the computation of scalar weights, and avoids the computation of matrix weights so that the computational burden can obviously be reduced. Based on this fusion criterion and Kalman predictor, an optimal information fusion filter for the input white noise, which can be applied to seismic data processing in oil exploration, is given for discrete time-varying linear stochastic control systems measured by multiple sensors with correlated noises. It has a two-layer fusion structure. The first fusion layer has a netted parallel structure to determine the first-step prediction error cross-covariance for the state and the filtering error cross-covariance for the input white noise between any two sensors at each time step. The second fusion layer is the fusion center to determine the optimal scalar weights and obtain the optimal fusion filter for the input white noise. Two simulation examples for Bernoulli–Gaussian white noise filter show the effectiveness.