传感技术学报2025，Vol.38Issue(9)：1606-1614,9.DOI:10.3969/j.issn.1004-1699.2025.09.009
自适应加权算术平均融合的分布式多伯努利目标跟踪
Adaptive Weighted Arithmetic Average Fusion for Distributed Multi-Bernoulli Target Tracking
摘要
Abstract
A distributed multi-Bernoulli target tracking algorithm based on adaptive weighted arithmetic average fusion is proposed to solve the problem that the fusion performance is degraded due to the inconsistency of sensor performance ignored by fixed fusion weights.Firstly,the detection probability time-varying model related to distance and the clutter probability time-varying model related to the number of clutter measurements are defined.Then,the sensor credibility depending on the detection probability and clutter probabil-ity is defined to measure the reliability of the data involved in the fusion in the sensor.Finally,local filtering is implemented in the framework of cardinality balanced multi-Bernoulli filter,and the adaptive fusion weight composed of credibility is introduced into the arithmetic average fusion of posterior densities,so as to realize the distributed multi-target tracking algorithm of adaptive weighted arith-metic average fusion.Simulation experiments show that,compared with fixed fusion weights,the tracking algorithm with the proposed adaptive fusion weight has higher estimation accuracy for target location and number as well as better fusion performance.关键词
目标跟踪/自适应加权融合/分布式算法/多伯努利滤波器/传感器可信度Key words
target tracking/adaptive weighted fusion/distributed algorithm/multi-Bernoulli filter/sensor credibility分类
信息技术与安全科学引用本文复制引用
陈泽铭,吴孙勇,韦春玲,郑翔飞..自适应加权算术平均融合的分布式多伯努利目标跟踪[J].传感技术学报,2025,38(9):1606-1614,9.基金项目
桂林电子科技大学数学与计算科学学院研究生创新项目(2022YJSCX02) （2022YJSCX02）
桂林电子科技大学研究生教育创新计划项目(2022YCXS145) （2022YCXS145）
