Distributed fusion method of multi-Bernoulli filter

Abstract

The invention discloses a distributed fusion method of a multi-Bernoulli filter, which comprises the steps of carrying out local filtering on sensor nodes in a sensor network by adopting the multi-Bernoulli filter, receiving multi-Bernoulli posterior distribution of adjacent nodes through a communication link, then realizing joint grouping of target tracks among sensors based on a generalized covariance intersection information entropy, and finally calculating parameters of distributed fusion of the multi-Bernoulli filter under each target track group so as to obtain fused posterior multi-Bernoulli distribution parameters. The distributed fusion method can efficiently realize distributed fusion of the multi-Bernoulli filter under the condition of a large quantity of targets, and has the advantages of small approximation error, low implementation complexity and the like.

Description

technical field [0001] The invention belongs to the field of multi-sensor fusion, and in particular relates to a multi-Bernoulli filter distributed fusion method. Background technique [0002] Single-sensor detection systems cannot achieve accurate and effective detection of time-varying multiple targets due to their single detection angle of view, single detection capability, and complex and diverse detection environments. A sensor network detection system based on the multi-directional scattering mechanism is urgently needed to Meet the requirements of stable and continuous detection tasks for multiple targets in complex detection scenarios. On the other hand, with the improvement of computing, storage, and communication capabilities of miniaturized detectors, the multi-sensor network system constructed by this type of detector is favored, and the system provides a platform for collecting diverse target information for multi-target detection. [0003] Multi-sensor and mul...

AI Technical Summary

Problems solved by technology

However, when the multi-Bernoulli filter-based GCI fusion analytical form mathematical expression realizes the fusion of multi-objective posterior distributions, there are disadvantages such as high cost of computing resources and complex fusion structure, which makes the unoptimized distributed fusion algorithm unable to meet Distributed sensor networks require high real-time processing of local node information and short delays in information transmission between nodes. Therefore, there is an urgent need for an unoptimized distributed fusion algorithm with robust performance, high efficiency, and an efficient implementation method with a simple fusion structure. , to approach or meet the actual needs of distributed real-time processing, is the key to apply the distributed multi-Bernoulli filter fusion technology to a wider range of practical application scenarios

Images