非高斯噪声下基于KL散度最小化的目标跟踪  

KLD Minimization-Based Target Tracking Under Non-Gaussian Noise

作　　者：霍勇进 周林[1] 陈赞如 苗天一 张前程[1] HUO Yongjin;ZHOU Lin;CHEN Zanru;MIAO Tianyi;ZHANG Qiancheng(Artificial Intelligence School,Henan University,Zhengzhou 450000,China)

机构地区：[1]河南大学人工智能学院,郑州450000

出　　处：《电光与控制》2024年第8期38-43,49,共7页Electronics Optics & Control

基　　金：河南大学研究生培养创新与质量提升行动计划项目(SYLKC 2022013,SYLYC2023188)。

摘　　要：在复杂环境下的目标跟踪系统中,由于受随机脉冲干扰、建模误差、未知异常值等因素影响,系统模型的过程噪声和测量噪声呈现出非高斯重尾的复杂特性。提出了一种在分布式融合框架下基于KL散度(KLD)最小化的算法。首先,包含了目标状态、过程噪声和测量噪声等多个参数的先验模型被构成学生t分布;其次,KLD最小化的方法解决近似分布拟合真实分布差距过大的问题,提高了学生t建模的准确性;最后,采用协方差交叉融合策略实现对局部平台状态估计融合与修正。仿真结果表明,所提算法较传统的NKF、STF、MCCKF算法,具有更高的估计精度。In the target tracking system in complex environment,due to the influence of random pulse interference,modeling error,unknown outliers and other factors,the process noise and measurement noise of the system model show complex non-Gaussian heavy-tailed characteristics.A method based on the KL Divergence(KLD)minimization in the distributed fusion framework is proposed.Firstly,a priori model including many parameters such as target state,process noise and measurement noise is constructed as a student't t distribution.Secondly,KLD minimization solves the problem of distance in fitting approximate distribution to the real distribution,and improving the accuracy of student't t modeling.Finally,the Covariance Intersection(CI)fusion strategy is adopted to realize the fusion and correction of local platform state estimation.The simulation results show that the proposed algorithm has higher estimation accuracy compared with the traditional NKF,STF and MCCKF algorithms.

关 键 词：非高斯重尾噪声 KL散度最小化 协方差交叉融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]