Optimal fusion estimation for stochastic systems with cross-correlated sensor noises

Abstract

This paper is concerned with the optimal fusion of sensors with cross-correlated sensor noises. By taking linear transformations to the measurements and the related parameters, new measurement models are established, where the sensor noises are decoupled. The centralized fusion with raw data, the centralized fusion with transformed data, and a distributed fusion estimation algorithm are introduced, which are shown to be equivalent to each other in estimation precision, and therefore are globally optimal in the sense of linear minimum mean square error (LMMSE). It is shown that the centralized fusion with transformed data needs lower communication requirements compared to the centralized fusion using raw data directly, and the distributed fusion algorithm has the best flexibility and robustness and proper communication requirements and computation complexity among the three algorithms (less communication and computation complexity compared to the existed distributed Kalman filtering fusion algorithms). An example is shown to illustrate the effectiveness of the proposed algorithms.

State estimation and data fusion for multirate sensor networks

“…They are not necessarily equivalent. An example is given in . When computational complexity is concerned, the OSF and OBF are comparable, but the OSF is more efficient because when time delay exists in collecting data, the OBF waits until all the sensors' information at a given time are collected, while the OSF handles data sequentially, and the estimation results have nothing to do with the order of the sensors to be fused; therefore, the fusion center can handle any sensor's information that are available.…”

Sequential fusion estimation for multisensor systems with non-Gaussian noises

“…The aim of multisensor fusion estimation is to make the best use of the local measurements or local estimators generated from every single sensor, to get the fusion estimation, which has higher accuracy than any local estimator that barely uses single sensor's information [1,3]. It is first studied in military applications and has been developed in many high-technology fields, such as aerospace, guidance, control, defense, navigation of intelligent vehicles, positioning of robotics, target tracking, monitoring and fault detection [1,[4][5][6][7].…”

Energy consumption diagnosis in the iron and steel industry via the Kalman filtering algorithm with a data-driven model

“…In literature, the statistical characteristics of the sample data are used to describe the system using the mean and the variance of the state variables when it is difficult to obtain the exact mathematical model. Yan et al [5] studies the optimal fusion of sensors data by taking linear transformations to establish a new measurement model to decouple noises. Sohlberg [6] dealt with condition monitoring and failure diagnosis for a steel strip rinsing process based on a priori knowledge and data.…”