Square Root Volumetric Filtering Method Based on Mixed Information Weighted Consistency

Abstract

The invention discloses a square root volumetric filtering method based on mixed information weighted consistency, which belongs to sensor information fusion technology and relates to the problem of distributed nonlinear state filtering. The filtering method first uses the square root volume rule to obtain the predicted information vector and the square root factor of the information matrix of each node for the target state; then calculates the corresponding local measurement information vector and the square root factor of the information matrix based on the observation information and state prediction information of each node; finally Through the consistent iteration of information exchange between neighbor nodes, the consistent state prediction information, measurement information and measurement standardization factor are obtained, and the estimated state of the target is updated through weighted fusion. The filtering method effectively solves the problem of node state overestimation under the condition of limited iterations, improves the consistent convergence speed and estimation accuracy of state information, and improves the numerical stability of the filtering method.

Description

technical field [0001] The invention belongs to sensor network information fusion technology, relates to the problem of distributed nonlinear state filtering, and provides a square root volumetric filtering method based on mixed information weighted consistency. Background technique [0002] For the multi-sensor target state estimation problem, the traditional method often adopts a centralized structure, which can obtain high-precision state estimation results, but requires high computing power of the fusion center, resulting in large computing and communication consumption of the system, and it is difficult to guarantee the estimation. real-time. In addition, the fusion center is very important to the entire estimation system. If it is damaged or fails, the system will be paralyzed, unable to perform state estimation tasks, and the scalability of the system is poor. Compared with the centralized structure, the distributed state estimation algorithm has many advantages such...

AI Technical Summary

Problems solved by technology

In addition, the movement of the target and the observation of the sensor are often non-linear, and the computing power and energy of each node in the network are also limited. How to design a stable and efficient distributed consensus state estimation algorithm in the case of limited network resources is a problem. Issues worthy of further study

Images