基于轨迹泊松多伯努利混合滤波器的浅海匹配场连续跟踪方法

周玉媛; 孙超; 谢磊

doi:10.7498/aps.72.20230124

摘要

匹配场跟踪方法依据模糊度函数时间序列中声源位置移动的连续性和伪峰位置的无序性, 可实现水下声源轨迹跟踪. 然而, 受到浅海空时起伏波导环境和声源复杂运动场景的双重影响, 已有匹配场跟踪方法易出现轨迹中断、交叉混叠和虚假轨迹等现象, 导致不连续的轨迹跟踪结果. 针对这一问题, 本文基于轨迹泊松多伯努利混合滤波器, 利用模糊度函数中峰值位置距离似然和峰值幅度似然的一致性, 提出一种匹配场连续跟踪方法. 该方法应用于SWellEx-96实验数据并由线性规划准则度量跟踪性能, 结果表明: 相比已有匹配场跟踪和基于随机有限集的多目标跟踪方法, 所提方法实现了两个水下运动声源轨迹连续跟踪, 其中, 轨迹状态随机有限集的建模方式以及在轨迹空间内执行预测和更新步骤, 可以防止在未持续发声且数量未知的声源跟踪过程中出现轨迹中断和混叠现象; 结合模糊度函数峰值位置和幅度信息执行数据关联步骤, 可抑制虚假轨迹.

关键词:

Abstract

In the shallow water waveguide, matched field tracking methods use the continuity of the peak position of the moving source and the disorder of pseudo-peaks on the sequential ambiguity surfaces to track the underwater source trajectory. However, owing to the dual influence of the space-time fluctuating shallow water waveguide environment and the complex sources motion scene, the existing matching field tracking methods are prone to track interruption, switches and false track phenomena, leading to discontinuous tracking results. Using the consistency between the peak position distance likelihood and the peak amplitude likelihood of sequential ambiguity surfaces, a continuous matched field tracking method is proposed based on the trajectory Poisson multi-Bernoulli mixture filter in this paper. The proposed method is applied to SWellEx-96 experimental data, and the tracking performance is measured by the linear programming metric. The results show that compared with the existing matching field tracking method and multi-target tracking method via random finite set, the proposed method achieves continuous tracking and accurate quantity estimation of moving sources trajectory. Among them, the prediction step and updating step in the trajectory space can avoid the phenomenon of trajectory interruption and switches in unvoiced periods.

Keywords:

作者及机构信息

1.
西北工业大学航海学院, 西安　710072

2.
西北工业大学, 陕西省水下信息技术重点实验室, 西安　710072

通信作者: 孙超, csun@nwpu.edu.cn

基金项目: 国家自然科学基金(批准号: 12274348, 11534009)资助的课题.

Authors and contacts

1.
School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, China

2.
Key Laboratory of Ocean Acoustic and Sensing, Ministry of Industry and Information Technology, Northwestern Polytechnical University, Xi’an 710072, China

Corresponding author: Sun Chao, csun@nwpu.edu.cn

Funds: Project supported by the National Natural Science Foundation of China (Grant Nos. 12274348, 11534009).

文章全文 : translate this paragraph

参考文献

[1]	Bucker H 1976 J. Acoust. Soc. Am. 59 368 Google Scholar
[2]	Bucker H 1994 J. Acoust. Soc. Am. 96 3809 Google Scholar
[3]	Fialkowski L T, Perkins J S, Collins M D 2001 J. Acoust. Soc. Am. 110 739 Google Scholar
[4]	Maranda B H, Fawcett J A 1991 IEEE J. Ocean. Eng. 16 189 Google Scholar
[5]	Fawcett J A, Maranda B H 1993 J. Acoust. Soc. Am. 94 1363 Google Scholar
[6]	Zala C A, Ozard J M, Wilmut M J 1998 J. Acoust. Soc. Am. 103 374 Google Scholar
[7]	Tantum S L, Nolte L W 2002 J. Acoust. Soc. Am. 112 119 Google Scholar
[8]	Mahler R P 2014 Advances in Statistical Multisource-Multitarget Information Fusion (Boston, London: Artech house) p83
[9]	Yardim C, Michalopoulou Z, Gerstoft P 2011 IEEE J. Ocean. Eng. 36 71 Google Scholar
[10]	Vo B N, Ma W K 2006 IEEE Trans. Signal. Proces. 54 4091 Google Scholar
[11]	Gruden P, White P R 2016 J. Acoust. Soc. Am. 140 1981 Google Scholar
[12]	Gruden P, White P R 2016 J. Acoust. Soc. Am. 148 3014 Google Scholar
[13]	Gruden P, Nosal E M 2021 J. Acoust. Soc. Am. 150 3399 Google Scholar
[14]	Kupilik M J, Petersen T 2014 J. Acoust. Soc. Am. 136 1736 Google Scholar
[15]	Georgescu R, Willett P 2012 IEEE J. Ocean. Eng. 37 220 Google Scholar
[16]	García-Fernández Á F, Williams J L 2018 IEEE Trans. Signal. Proces. 54 1883 Google Scholar
[17]	García-Fernández Á F, Svensson L 2020 IEEE Trans. Signal. Proces. 68 4933 Google Scholar
[18]	Vo B T, Vo B N 2013 IEEE Trans. Signal. Proces. 61 3460 Google Scholar
[19]	Tracey B H 2005 J. Acoust. Soc. Am. 118 1372 Google Scholar
[20]	Lerro D, Bar-Shalom Y 1993 IEEE Trans. Aerosp. Electron. Syst. 29 404 Google Scholar
[21]	Bendat J S, Piersol A G 2010 Random Data: Analysis and Measurement Procedures (4th Ed.) (Hoboken, NJ: Wiley) p604
[22]	Murray J, Ensberg D The SWellEx-96 experiment http://swellex96.ucsd.edu/ (Last viewed July 2021
[23]	Porter M B 1991 The KRAKEN Normal Mode Program (La Spezia: SACLANT Undersea Research Centre
[24]	Gemba K L, Nannuru S 2017 J. Acoust. Soc. Am. 141 3411 Google Scholar
[25]	Booth N O, Baxley P A 1996 IEEE J. Ocean. Eng. 21 402 Google Scholar
[26]	García-Fernández Á F, Rahmathullah A S 2020 IEEE Trans. Signal. Proces. 68 3917 Google Scholar

施引文献

图 1 SWellEx-96测线S5实验　(a)发射船轨迹; (b)波导环境示意图

Fig. 1. The SWellEx-96 event S5: (a) The launch ship track; (b) SWellEx-96 waveguide.

下载: 全尺寸图片幻灯片

图 2 双目标距离维轨迹　(a) 现有匹配场跟踪方法轨迹估计结果; (b) 轨迹真值

Fig. 2. Range dimension trajectory of two targets: (a) MFT result; (b) the truth.

下载: 全尺寸图片幻灯片

图 3 匹配场跟踪性能影响因素　(a)沿轨迹的水深变化; (b)平均阵元接收信噪比; (c)相关系数

Fig. 3. Influencing factors of matching field tracking performance: (a) The bottom depth along the trajectory; (b) average element level SNR; (c) correlation coefficient.

下载: 全尺寸图片幻灯片

图 4 所有时间步测量集合　(a) 距离维空间分布; (b) 元素数量

Fig. 4. Measurements sets of all steps: (a) Spatial distribution of measurement elements in range dimension; (b) the number of elements in the measurement set.

下载: 全尺寸图片幻灯片

图 5 四种方法的距离维轨迹估计结果　(a) TPMBM-A-MFT; (b) TPMBM-MFT; (c) PMBM-MFT; (d) GMPHD-MFT

Fig. 5. Range dimension trajectory estimation results for four methods: (a) TPMBM-A-MFT; (b) TPMBM-MFT; (c) PMBM-MFT; (d) GMPHD-MFT.

下载: 全尺寸图片幻灯片

图 6 TPMBM-A-MFT的跟踪过程　(a)第6.29 min; (b)第6.77 min; (c)第11.13 min; (d)第11.61 min; (e)第21.77 min; (f)第22.26 min; (g)第42.10 min; (h)第54.19 min; (i)第60.00 min

Fig. 6. Tracking process of TPMBM-A-MFT method: (a) 6.29 minutes; (b) 6.77 minutes; (c) 11.13 minutes; (d) 11.61 minutes; (e) 21.77 minutes; (f) 22.26 minutes; (g) 42.10 minutes; (h) 54.19 minutes; (i) 60.00 minutes.

下载: 全尺寸图片幻灯片

图 7 四种跟踪方法在每个时间步的LP度量结果

Fig. 7. LP metrics at each time step for four methods.

下载: 全尺寸图片幻灯片

图 8 LP的4种分解度量　(a)匹配位置误差; (b)轨迹缺失代价; (c)虚假轨迹代价; (d)轨迹混叠代价

Fig. 8. Four decomposition metrics of LP: (a) Matched localization error; (b) missed trajectory cost; (c) false trajectory cost; (d) switching cost.

下载: 全尺寸图片幻灯片

表 1 TPMBM-A-MFT算法流程

Table 1. Steps of TPMBM-A-MFT algorithm.

步骤1　建立匹配场轨迹状态空间模型　　　　变换模糊度函数: (26)式; 　　　　给出状态方程和观测方程: (27)式和(28)式;
步骤2　根据(27)式对$ p({{\bf X}_k}) $参数集预测: 2.2.1节;
步骤3　根据(28)式对$ p({{\bf X}_k}) $参数集更新: 2.2.2节;
步骤4　结合峰值位置与幅度实现数据关联: (36)式;
步骤5　估计轨迹状态和数量: (25)式.
注: 初始时间步中$ {\bf{X} }_{k = 1}^{\rm{d}} = \varnothing $且泊松强度$ \lambda _{k = 1}^{\rm{u}} = \lambda _{k = 1}^{\rm{b}} $.

下载: 导出CSV

表 2 匹配场跟踪过程的滤波参数设置

Table 2. Filter parameters setup for matched field tracking process.

${q_r}$	${q_d}$	${\sigma _r}$	${\sigma _d}$	$T$	${w^{\text{b}}}$	${P^{\text{S}}}$	${P^{\text{D}}}$	$\varGamma $
$6 \times {10^{ - 3}}$	$6 \times {10^{ - 3}}$	$0.05$	$2$	$1$	$5 \times {10^{ - 3}}$	$0.99$	$\{ 0.2, 0.5\} $	$0.1$

下载: 导出CSV

[1]	Bucker H 1976 J. Acoust. Soc. Am. 59 368 Google Scholar
[2]	Bucker H 1994 J. Acoust. Soc. Am. 96 3809 Google Scholar
[3]	Fialkowski L T, Perkins J S, Collins M D 2001 J. Acoust. Soc. Am. 110 739 Google Scholar
[4]	Maranda B H, Fawcett J A 1991 IEEE J. Ocean. Eng. 16 189 Google Scholar
[5]	Fawcett J A, Maranda B H 1993 J. Acoust. Soc. Am. 94 1363 Google Scholar
[6]	Zala C A, Ozard J M, Wilmut M J 1998 J. Acoust. Soc. Am. 103 374 Google Scholar
[7]	Tantum S L, Nolte L W 2002 J. Acoust. Soc. Am. 112 119 Google Scholar
[8]	Mahler R P 2014 Advances in Statistical Multisource-Multitarget Information Fusion (Boston, London: Artech house) p83
[9]	Yardim C, Michalopoulou Z, Gerstoft P 2011 IEEE J. Ocean. Eng. 36 71 Google Scholar
[10]	Vo B N, Ma W K 2006 IEEE Trans. Signal. Proces. 54 4091 Google Scholar
[11]	Gruden P, White P R 2016 J. Acoust. Soc. Am. 140 1981 Google Scholar
[12]	Gruden P, White P R 2016 J. Acoust. Soc. Am. 148 3014 Google Scholar
[13]	Gruden P, Nosal E M 2021 J. Acoust. Soc. Am. 150 3399 Google Scholar
[14]	Kupilik M J, Petersen T 2014 J. Acoust. Soc. Am. 136 1736 Google Scholar
[15]	Georgescu R, Willett P 2012 IEEE J. Ocean. Eng. 37 220 Google Scholar
[16]	García-Fernández Á F, Williams J L 2018 IEEE Trans. Signal. Proces. 54 1883 Google Scholar
[17]	García-Fernández Á F, Svensson L 2020 IEEE Trans. Signal. Proces. 68 4933 Google Scholar
[18]	Vo B T, Vo B N 2013 IEEE Trans. Signal. Proces. 61 3460 Google Scholar
[19]	Tracey B H 2005 J. Acoust. Soc. Am. 118 1372 Google Scholar
[20]	Lerro D, Bar-Shalom Y 1993 IEEE Trans. Aerosp. Electron. Syst. 29 404 Google Scholar
[21]	Bendat J S, Piersol A G 2010 Random Data: Analysis and Measurement Procedures (4th Ed.) (Hoboken, NJ: Wiley) p604
[22]	Murray J, Ensberg D The SWellEx-96 experiment http://swellex96.ucsd.edu/ (Last viewed July 2021
[23]	Porter M B 1991 The KRAKEN Normal Mode Program (La Spezia: SACLANT Undersea Research Centre
[24]	Gemba K L, Nannuru S 2017 J. Acoust. Soc. Am. 141 3411 Google Scholar
[25]	Booth N O, Baxley P A 1996 IEEE J. Ocean. Eng. 21 402 Google Scholar
[26]	García-Fernández Á F, Rahmathullah A S 2020 IEEE Trans. Signal. Proces. 68 3917 Google Scholar

[1]	汪磊, 黄益旺, 郭霖, 任超. 浅海粗糙海底声散射建模及声场特性. 物理学报, 2024, 73(3): 034301. doi: 10.7498/aps.73.20231472
[2]	韦宜政, 孙超, 朱启轩. 浅海矢量声场极化特性的深度分布规律. 物理学报, 2024, 73(9): 094302. doi: 10.7498/aps.73.20231767
[3]	李永飞, 郭瑞明, 赵航芳. 浅海内波环境下声场干涉条纹的稀疏重建. 物理学报, 2023, 72(7): 074301. doi: 10.7498/aps.72.20221932
[4]	周玉媛, 孙超, 谢磊, 刘宗伟. 基于波束-波数域非相干匹配的浅海运动声源深度估计方法. 物理学报, 2023, 72(8): 084302. doi: 10.7498/aps.72.20222361
[5]	陶建飞, 夏勤智, 廖临谷, 刘杰, 刘小井. 强激光场原子电离光电子轨迹干涉全息理论及应用. 物理学报, 2022, 71(23): 233206. doi: 10.7498/aps.71.20221296
[6]	王宣, 孙超, 李明杨, 张少东. 不确定浅海环境中水平阵角度域子空间检测. 物理学报, 2022, 71(8): 084304. doi: 10.7498/aps.71.20211742
[7]	张少东, 孙超, 谢磊, 刘雄厚, 王宣. 浅海波导环境不确定性对声源功率估计的影响. 物理学报, 2021, 70(24): 244301. doi: 10.7498/aps.70.20210852
[8]	周曜智, 李春, 李晨阳, 李清廉. 超声速横向气流中液体射流的轨迹预测与连续液柱模型. 物理学报, 2020, 69(23): 234702. doi: 10.7498/aps.69.20200903
[9]	钱治文, 商德江, 孙启航, 何元安, 翟京生. 三维浅海下弹性结构声辐射预报的有限元-抛物方程法. 物理学报, 2019, 68(2): 024301. doi: 10.7498/aps.68.20181452
[10]	孟瑞洁, 周士弘, 李风华, 戚聿波. 浅海波导中低频声场干涉简正模态的判别. 物理学报, 2019, 68(13): 134304. doi: 10.7498/aps.68.20190221
[11]	李晓曼, 张明辉, 张海刚, 朴胜春, 刘亚琴, 周建波. 一种基于模态匹配的浅海波导中宽带脉冲声源的被动测距方法. 物理学报, 2017, 66(9): 094302. doi: 10.7498/aps.66.094302
[12]	林呈, 张华堂, 盛志浩, 余显环, 刘鹏, 徐竟文, 宋晓红, 胡师林, 陈京, 杨玮枫. 用推广的量子轨迹蒙特卡罗方法研究强场光电子全息. 物理学报, 2016, 65(22): 223207. doi: 10.7498/aps.65.223207
[13]	戚聿波, 周士弘, 张仁和. 浅海波导中折射类简正波的warping变换. 物理学报, 2016, 65(13): 134301. doi: 10.7498/aps.65.134301
[14]	张同伟, 杨坤德. 一种水平变化波导中匹配场定位的虚拟时反实现方法. 物理学报, 2014, 63(21): 214303. doi: 10.7498/aps.63.214303
[15]	张同伟, 杨坤德, 马远良, 黎雪刚. 浅海中水平线列阵深度对匹配场定位性能的影响. 物理学报, 2010, 59(5): 3294-3301. doi: 10.7498/aps.59.3294
[16]	余赟, 惠俊英, 陈阳, 孙国仓, 滕超. 浅海低频声场中目标深度分类方法研究. 物理学报, 2009, 58(9): 6335-6343. doi: 10.7498/aps.58.6335
[17]	郑春兰, 李同保, 马艳, 马珊珊, 张宝武. 激光驻波场中Cr原子运动轨迹与汇聚沉积的分析. 物理学报, 2006, 55(9): 4528-4534. doi: 10.7498/aps.55.4528
[18]	张家树, 肖先赐. 连续混沌系统的非线性自适应预测跟踪控制. 物理学报, 2001, 50(11): 2092-2096. doi: 10.7498/aps.50.2092
[19]	陈仁术, 西门纪业. 扇形重叠场中三级离子轨迹理论(Ⅰ)——轨迹计算和矩阵表示. 物理学报, 1982, 31(6): 722-737. doi: 10.7498/aps.31.722
[20]	张仁和. 浅海表面声道中的简正波声场. 物理学报, 1975, 24(3): 200-209. doi: 10.7498/aps.24.200

计量

文章访问数: 1557
PDF下载量: 44
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板