基于LMB滤波器高斯混合实现的传感器控制方法  

Research on Sensor Control Method Based on Gaussian Mixture of LMB Filter

作　　者：廖晨 LIAO Chen(College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050)

机构地区：[1]兰州理工大学电气工程与信息工程学院,兰州730050

出　　处：《舰船电子工程》2022年第7期37-43,共7页Ship Electronic Engineering

基　　金：国家自然科学基金项目(编号:61873116,51668039,61763029);国防基础科研项目(编号:JCKY2018427C002)资助。

摘　　要：针对多目标跟踪优化过程中的传感器控制问题,论文基于标签多伯努利(LMB)滤波器的高斯混合实现方式提出了两种传感器控制策略。首先,简单阐述了LMB滤波器的高斯混合实现过程,采用高斯分量去近似多目标概率密度,通过求解两个高斯分布之间的柯西施瓦茨(CS)散度,得到以信息增益最大化为准则的传感器控制方案;其次,利用多伯努利参数替代标签多伯努利参数进行“去标签”的伪更新步骤,有效降低了计算复杂度;此外,借助战术重要性评估指标(TSM)函数,提出一种基于目标威胁度的传感器控制算法;最后,合理设定仿真场景,对比验证了两种所提算法的有效性和实用性。Aiming at the sensor control problem in multi-target tracking optimization,two sensor control strategies are pro⁃posed based on gaussian mixture implementation of label multi-bernoulli(LMB)filter.Firstly,the Gaussian Mixture realization pro⁃cess of the LMB filter is briefly described.The Gaussian component is used to approximate the multi-target probability density.By solving the Cauchy-Schwartz(CS)divergence between the two Gaussian distributions,the sensor control scheme based on the maxi⁃mization of information gain is obtained.Secondly,the multi-bernoulli parameter is used to replace the label multi-bernoulli param⁃eter for the pseudo-update step of"de-label",which effectively reduces the computational complexity.In addition,with the help of TSM function,a sensor control algorithm based on target threat degree is proposed.Finally,the simulation scenario is set reason⁃ably,and the effectiveness and practicability of the two algorithms are compared and verified.

关 键 词：多目标跟踪 标签多伯努利滤波器 柯西施瓦茨散度 传感器控制 目标威胁度 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]