Page 89 - 《应用声学》2021年第3期

第 40 卷 第 3 期 Vol. 40, No. 3
2021 年 5 月 Journal of Applied Acoustics May, 2021

⋄ 研究报告 ⋄

基于粒子滤波的多普勒信息辅助目标

定位跟踪算法

张 蒙 1,2 王海斌 1† 张海如 1 汪 俊 1 胡治国 1

(1 中国科学院声学所 声场声信息国家重点实验室 北京 100190)
(2 中国科学院大学 北京 100049)
摘要：多基地声呐探测系统主要通过测量回波的时延和方位信息进行目标定位与跟踪，定位精度受声速、时延
和方位测量误差的影响较大，可以通过多普勒信息辅助进一步提高定位跟踪精度。现有的多普勒信息辅助定
位跟踪算法多适用于单基地声呐系统，多基地中的多普勒测量值与目标状态的关系更为复杂，需要研究新的
融合方法。该文提出了一种适用于多基地声呐系统的多普勒信息辅助采样重要性重采样目标定位跟踪算法，
将多普勒信息融入到粒子滤波的重采样过程，使重采样后的粒子集合更逼近目标的真实状态分布，从而提高
了目标定位跟踪精度。数值仿真实验结果表明，提出的目标定位跟踪算法可以有效融合多普勒信息，提升目标
定位跟踪精度。
关键词：多基地主动声呐；多普勒频移；粒子滤波；目标定位与跟踪
中图法分类号: TB566 文献标识码: A 文章编号: 1000-310X(2021)03-0407-08
DOI: 10.11684/j.issn.1000-310X.2021.03.012

Doppler information-assisted target tracking algorithm based on particle filter

ZHANG Meng 1,2 WANG Haibin 1 ZHANG Hairu 1 WANG Jun 1 HU Zhiguo 1
(1 State Key Laboratory of Acoustics, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China)
(2 University of Chinese Academy of Sciences, Beijing 100049, China)

Abstract: The multistatic active sonar system mainly locates and tracks the target by measuring the time
delay and azimuth information of the echo. The locating accuracy is greatly affected by the error of sound
speed, time delay and azimuth measurement, and can be further improved by Doppler information assistance.
However, existing Doppler information-assisted locating and tracking algorithms are mostly suitable for mono-
static sonar systems. The relationship between Doppler shift and target states in multistatic active sonar
system is more complicated and new fusion methods are required. In this paper, the Doppler information as-
sisted target locating and tracking method applicable to the multistatic sonar system is studied, and a Doppler
information-assisted sampling importance resampling (SIR) target locating and tracking algorithm is proposed.
The algorithm incorporates Doppler information into the resampling process of particle filter, so that the resam-
pled particle set is closer to the true state distribution of the target, thereby improving the accuracy of target
locating and tracking. The results of numerical simulation experiments show that the proposed algorithm can
effectively fuse Doppler information to improve the target locating and tracking accuracy.
Keywords: Multistatic active sonar; Doppler shift; Particle filter; Target locating and tracking

2020-07-09 收稿; 2020-10-23 定稿
作者简介: 张蒙 (1991– ), 男, 山东临沂人, 博士研究生, 研究方向: 多基地主动探测、声呐信号处理。
通信作者 E-mail: whb@mail.ioa.ac.cn
†