搜索

x

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于迁移学习的水下目标定位方法仿真研究

雷波 何兆阳 张瑞

引用本文:
Citation:

基于迁移学习的水下目标定位方法仿真研究

雷波, 何兆阳, 张瑞

Simulation study of underwater intruder localization based on transfer learning

Lei Bo, He Zhao-Yang, Zhang Rui
PDF
HTML
导出引用
  • 针对水下前向散射探测中基于敏感核函数的定位方法存在环境失配带来的稳健性问题, 提出了一种基于迁移学习的前向散射定位方法, 利用模型生成ϕ的目标前向散射声场扰动信息训练卷积神经网络, 将目标定位问题转化为分类问题. 在基于先验信息和仿真数据集的预训练模型基础上, 通过少量实验数据集对神经网络参数进行迁移学习, 以提高神经网络模型的稳健性. 仿真结果表明, 该方法在声速剖面失配下可以实现对目标较准确的定位, 且对目标散射函数、海底底质、阵元数和布设深度等参数不甚敏感, 方法具有较好的稳健性.
    Forward scattering of the target could cause the amplitude and phase aberration of the received sound field, which received attentions in harbor monitoring and anti-submarine. However, the localization under forward scattering configuration is a challenging task due to the strong direct blast. The method based on sensitive kernel function which exploit the aberration of the received signals is sensitive to the environment mismatch and a localization method based on transfer learning framework is developed. The envelopes of aberrations caused by the forward scattering of intruder are firstly extracted by applying pulse compression technique on the received signals, and then normalized by comparing with the case of intruder absent. The data set near the first arrivals on the normalized aberrations are selected as the learning physical parameters. A convolution neural network is trained with these data generated by the forward scattering model to establish a mapping relationship between intruder’s localization and the aberrations of received signal, thus the localization problem is transformed into classification. In the second step, the parameters of the convolutional pooling layer in the pre-trained model are frozen in the transfer learning procedure, and the parameters of the fully connected layer in the pre-trained model are updated using a small amount of data under the fluctuated environment. Simulation of the localization of ellipsoidal targets with a signal-to-noise ratio of 0 dB under a shallow water environment is performed for a scenario to explore the robustness of the method. The results show that the accurate target localization could be achieved in the case of sound velocity profile mismatch. Also, the method is not significantly sensitive to the target scattering function, sound properties of sediment and deployment of transceivers. The sensitivities to the waveguide amplitude and phase fluctuations are further modeled. The results show that good localization accuracy can be obtained in a relatively stable environment, and results are distinguished between the presence and absence of the target. Since the proposed method is derived by the model and real data, the accurate scattering model and sufficient training data are not significantly necessary. The method may provide a promising way for forward scattering detection.
      通信作者: 雷波, lei.bo@nwpu.edu.cn
    • 基金项目: 国家自然科学基金(批准号: 61571366)资助的课题.
      Corresponding author: Lei Bo, lei.bo@nwpu.edu.cn
    • Funds: Project supported by the National Natural Science Foundation of China (Grant No. 61571366)
    [1]

    Coraluppi S 2006 IEEE J. Oceanic Eng. 31 964Google Scholar

    [2]

    Bekkerman I, Tabrikian J 2006 IEEE Trans. Signal Process. 54 3873Google Scholar

    [3]

    Kim S, Ku B, Hong W, Ko H 2008 IEEE Trans. Aerosp. Electron. Syst. 44 1371Google Scholar

    [4]

    Gillespie B, Rolt K, Edelson G 1997 Acoust. Imaging 23 501

    [5]

    Zverev V A, Matveev A L, Mityugov V V 1995 Acoust. Phys. 41 518

    [6]

    Zverev V A, Korotin P I, Matveev A L, Mityugov V V, Orlov D A, Salin B M, Turchin V I 2001 Acoust. Phys. 47 184Google Scholar

    [7]

    Song H, Kuperman W A, Akal T, Guerrini P 2003 IEEE J. Oceanic Eng. 28 246Google Scholar

    [8]

    Roux P, Kuperman W A, Hogkiss W S, Song H J, Akal T 2004 J. Acoust. Soc. Am. 116 1009Google Scholar

    [9]

    马敬广, 段敬伟, 惠俊英 2007 应用声学 26 135Google Scholar

    Ma J G, Duan J W, Hui J Y 2007 Applied Acoustics 26 135Google Scholar

    [10]

    Lei B, Yang K D, Ma Y L 2012 J. Acoust. Soc. Am. 132 284Google Scholar

    [11]

    Folegot T, Martinelli G, Guerrini P, Stevenson J M 2008 J. Acoust. Soc. Am. 124 2852Google Scholar

    [12]

    Marandet C, Roux P, Nicolas B, Mars J 2011 J. Acoust. Soc. Am. 129 85Google Scholar

    [13]

    Yildiz S, Roux P, Rakotonarivo S T, Marandet C, Kuperman W A 2014 J. Acoust. Soc. Am. 135 1800Google Scholar

    [14]

    Niu H Q, Reeves E, Gerstoft P 2017 J. Acoust. Soc. Am. 142 1176Google Scholar

    [15]

    Niu H Q, Ozanich E, Gerstoft P 2017 J. Acoust. Soc. Am. 142 455Google Scholar

    [16]

    Niu H Q, Gong Z X, Ozanich E, Gerstoft P, Wang H B, Li Z L 2019 J. Acoust. Soc. Am. 146 211Google Scholar

    [17]

    Huang Z Q, Xu J, Gong Z X, Wang H B, Yan Y H 2018 J. Acoust. Soc. Am. 143 2922Google Scholar

    [18]

    Liu Y N, Niu H Q, Li Z L 2019 Chin. Phys. Lett. 36 47Google Scholar

    [19]

    Liu W X, Yang Y X, Xu M Q, Lü L G, Liu Z W, Shi Y 2020 J. Acoust. Soc. Am 147 314Google Scholar

    [20]

    Pan S J, Yang Q 2010 IEEE Trans. Knowl. Data E 22 1345Google Scholar

    [21]

    Ingento F 1987 J. Acoust. Soc. Am. 82 2051Google Scholar

    [22]

    陈燕, 汤渭霖, 范威 2010 声学学报 3 35Google Scholar

    Chen Y, Tang W L, Fan W 2010 Acta Acustica 3 35Google Scholar

    [23]

    雷波 2018 水中目标前向散射声场特征及其应用 (北京: 科学出版社) 第58−60页

    Lei B 2018 Sound Field Characteristics of Forward Scattering of Target in Water and Its Application (Beijing: Science Press) pp58−60 (in Chinese)

    [24]

    Nair V, Hinton G E 2010 Proceedings of the 27th International Conference on Machine Learning Haifa, Israel, June 21—24, 2010 p807

    [25]

    Ye Z, Hoskinson E, Dewey R K 1997 J. Acoust. Soc. Am. 102 1964Google Scholar

    [26]

    Porter M B, Bucker H P 1987 J. Acoust. Soc. Am. 82 1349Google Scholar

  • 图 1  声屏障示意图

    Fig. 1.  Schematic diagram of sound barrier with transmitter and receiver arrays.

    图 2  定位方法流程图

    Fig. 2.  Flow chart of positioning method.

    图 3  卷积神经网络结构示意图

    Fig. 3.  Structure diagram of convolution neural network.

    图 4  仿真实验环境 (a) 仿真实验示意图; (b) 接收到的信号波形; (c) 声场扰动量$ A(t)$

    Fig. 4.  Simulation experiment: (a) Diagram of simulation experiment; (b) received signal waveforms; (c) sound field aberration $ A(t)$

    图 5  预训练模型的训练过程和预测结果 (a) 准确率变化; (b) 损失函数变化; (c) 无失配时的预训练模型预测结果

    Fig. 5.  Training process and prediction results of the pre-training model: (a) Variation of accuracy; (b) variation of loss function; (c) prediction results of the pre-training model without mismatch

    图 6  环境失配时神经网络模型预测结果 (a) 预训练模型预测结果; (b) 迁移学习后预测结果

    Fig. 6.  Prediction results of neural network model with environment mismatch: (a) Prediction results of pre-training model; (b) prediction results after transfer learning

    图 7  环境失配时敏感核函数预测结果 (a) 单个样本定位结果; (b)多样本预测结果

    Fig. 7.  Prediction results of sensitive kernel function with environmental mismatch: (a) Location results of single sample; (b) results of multiple samples

    图 8  定位准确率随起伏变化 (a) 定位准确率随幅度起伏的变化; (b) 定位准确率随相位起伏的变化

    Fig. 8.  Position accuracy with fluctuation variation: (a) Position accuracy with magnitude fluctuation; (b) position accuracy with phase fluctuation

    图 9  起伏信道下无目标定位预测结果 (a) 无目标幅度起伏$ \sigma/{\rm {max}}(h)=0.6$的定位预测结果; (b) 无目标相位起伏$ \phi=2\pi/3$的定位预测结果

    Fig. 9.  Position result without target in fluctuated channel: (a) Position results without target in magnitude fluctuated channel $ \sigma/{\rm {max}}(h)=0.6$; (b) position results without target in phase fluctuated channel $ \phi=2\pi/3$

    表 1  卷积神经网络参数设置

    Table 1.  Parameter setting of convolutional neural network

    结构参数具体设置
    池化方法最大池化
    优化器Adam
    损失函数交叉熵
    学习率0.001
    b随机失活层丢弃率0.5
    下载: 导出CSV

    表 2  目标散射函数失配时的仿真结果

    Table 2.  Simulation results of target scattering function mismatch

    实际目标
    长度/m
    实际目标圆柱半径/m目标失配时预测准确率/%迁移学习后预测准确率/%
    352.577.295.2
    40379.496.0
    45 3.5 74.8 93.6
    下载: 导出CSV

    表 3  海底底质失配时的仿真结果

    Table 3.  Simulation results of sediment properties mismatch

    实际海底
    底质类型
    密度/
    (g·cm–3)
    声速/
    (m·s–1)
    海底底质失配时
    预测准确率/%
    迁移学习后
    预测准确率/%
    泥砂1.806166878.496.0
    细砂1.957175374.693.6
    粗砂 2.034 183671.0 92.0
    下载: 导出CSV

    表 4  不同发射阵元数时的仿真结果

    Table 4.  Simulation results of different number of transmitting array elements

    发射阵元数无失配时预测准确率/%目标失配时预测准确率/%迁移学习后预测准确率/%
    395.877.689.4
    497.381.292.6
    5 98.7 83.4 95.0
    6 99.0 85.4 95.8
    下载: 导出CSV

    表 5  不同接收阵元数时的仿真结果

    Table 5.  Simulation results with different number of receiving array elements

    接收阵元数无失配时预测准确率/%目标失配时预测准确率/%迁移学习后预测准确率/%
    1792.671.490.8
    1994.677.292.2
    2198.783.4 95.0
    2398.986.8 96.6
    下载: 导出CSV

    表 6  不同布设深度时的仿真结果

    Table 6.  Simulation results of different layout depths

    发射阵元布设深度/m无失配时预测准确率/%目标失配时预测准确率/%迁移学习后预测准确率/%
    20—3296.281.490.8
    20—8098.783.495.0
    68—8096.884.0 91.2
    下载: 导出CSV
    Baidu
  • [1]

    Coraluppi S 2006 IEEE J. Oceanic Eng. 31 964Google Scholar

    [2]

    Bekkerman I, Tabrikian J 2006 IEEE Trans. Signal Process. 54 3873Google Scholar

    [3]

    Kim S, Ku B, Hong W, Ko H 2008 IEEE Trans. Aerosp. Electron. Syst. 44 1371Google Scholar

    [4]

    Gillespie B, Rolt K, Edelson G 1997 Acoust. Imaging 23 501

    [5]

    Zverev V A, Matveev A L, Mityugov V V 1995 Acoust. Phys. 41 518

    [6]

    Zverev V A, Korotin P I, Matveev A L, Mityugov V V, Orlov D A, Salin B M, Turchin V I 2001 Acoust. Phys. 47 184Google Scholar

    [7]

    Song H, Kuperman W A, Akal T, Guerrini P 2003 IEEE J. Oceanic Eng. 28 246Google Scholar

    [8]

    Roux P, Kuperman W A, Hogkiss W S, Song H J, Akal T 2004 J. Acoust. Soc. Am. 116 1009Google Scholar

    [9]

    马敬广, 段敬伟, 惠俊英 2007 应用声学 26 135Google Scholar

    Ma J G, Duan J W, Hui J Y 2007 Applied Acoustics 26 135Google Scholar

    [10]

    Lei B, Yang K D, Ma Y L 2012 J. Acoust. Soc. Am. 132 284Google Scholar

    [11]

    Folegot T, Martinelli G, Guerrini P, Stevenson J M 2008 J. Acoust. Soc. Am. 124 2852Google Scholar

    [12]

    Marandet C, Roux P, Nicolas B, Mars J 2011 J. Acoust. Soc. Am. 129 85Google Scholar

    [13]

    Yildiz S, Roux P, Rakotonarivo S T, Marandet C, Kuperman W A 2014 J. Acoust. Soc. Am. 135 1800Google Scholar

    [14]

    Niu H Q, Reeves E, Gerstoft P 2017 J. Acoust. Soc. Am. 142 1176Google Scholar

    [15]

    Niu H Q, Ozanich E, Gerstoft P 2017 J. Acoust. Soc. Am. 142 455Google Scholar

    [16]

    Niu H Q, Gong Z X, Ozanich E, Gerstoft P, Wang H B, Li Z L 2019 J. Acoust. Soc. Am. 146 211Google Scholar

    [17]

    Huang Z Q, Xu J, Gong Z X, Wang H B, Yan Y H 2018 J. Acoust. Soc. Am. 143 2922Google Scholar

    [18]

    Liu Y N, Niu H Q, Li Z L 2019 Chin. Phys. Lett. 36 47Google Scholar

    [19]

    Liu W X, Yang Y X, Xu M Q, Lü L G, Liu Z W, Shi Y 2020 J. Acoust. Soc. Am 147 314Google Scholar

    [20]

    Pan S J, Yang Q 2010 IEEE Trans. Knowl. Data E 22 1345Google Scholar

    [21]

    Ingento F 1987 J. Acoust. Soc. Am. 82 2051Google Scholar

    [22]

    陈燕, 汤渭霖, 范威 2010 声学学报 3 35Google Scholar

    Chen Y, Tang W L, Fan W 2010 Acta Acustica 3 35Google Scholar

    [23]

    雷波 2018 水中目标前向散射声场特征及其应用 (北京: 科学出版社) 第58−60页

    Lei B 2018 Sound Field Characteristics of Forward Scattering of Target in Water and Its Application (Beijing: Science Press) pp58−60 (in Chinese)

    [24]

    Nair V, Hinton G E 2010 Proceedings of the 27th International Conference on Machine Learning Haifa, Israel, June 21—24, 2010 p807

    [25]

    Ye Z, Hoskinson E, Dewey R K 1997 J. Acoust. Soc. Am. 102 1964Google Scholar

    [26]

    Porter M B, Bucker H P 1987 J. Acoust. Soc. Am. 82 1349Google Scholar

  • [1] 刘凯越, 李腾耀, 郑娜娥, 田志富, 蔡通, 王彦朝, 曹超华. 基于迁移学习的共形超构表面散射场高效智能计算方法.  , 2024, 73(23): 234101. doi: 10.7498/aps.73.20241160
    [2] 冯波, 徐文君, 蔡杰雄, 吴如山, 王华忠. 标量声波方程前向散射场的保相位理论及其线性化近似.  , 2023, 72(15): 159101. doi: 10.7498/aps.72.20230194
    [3] 孙涛, 袁健美. 基于迁移学习的钙钛矿材料带隙预测.  , 2023, 72(21): 218901. doi: 10.7498/aps.72.20231027
    [4] 杨玉莲, 刘黎明, 邓庆雪, 贾新鸿, 梁文燕, 姜利, 宋伟杰, 牟欣扬. 非线性效应对前向受激布里渊散射分布式传感的影响.  , 2022, 71(15): 154206. doi: 10.7498/aps.71.20220313
    [5] 王玉龙, 张晓虹, 李丽丽, 高俊国, 郭宁, 程成. 基于超声波声压衰减效应的局部放电源定位与强度标定.  , 2021, 70(9): 095209. doi: 10.7498/aps.70.20201727
    [6] 王身云, 郑淏予, 李阳. 近场目标自适应聚焦天线.  , 2020, 69(21): 218402. doi: 10.7498/aps.69.20201525
    [7] 高文, 汤洋, 朱明. 目标跟踪中目标模型更新问题的半监督学习算法研究.  , 2015, 64(1): 014205. doi: 10.7498/aps.64.014205
    [8] 张同伟, 杨坤德, 马远良, 汪勇. 一种基于单水听器宽带信号自相关函数的水下目标定位稳健方法.  , 2015, 64(2): 024303. doi: 10.7498/aps.64.024303
    [9] 刘宗伟, 孙超, 向龙凤, 易锋. 不确定海洋环境中的模态子空间重构稳健定位方法.  , 2014, 63(3): 034304. doi: 10.7498/aps.63.034304
    [10] 刘辉, 杨俊安, 王一. 基于流形学习的声目标特征提取方法研究.  , 2011, 60(7): 074302. doi: 10.7498/aps.60.074302
    [11] 伍怀龙, 田东风, 郝樊华, 龚建. 钚气溶胶环境中惰性气体氪迁移过程研究.  , 2011, 60(3): 032801. doi: 10.7498/aps.60.032801
    [12] 刘厚通, 陈良富, 苏林. Fernald前向积分用于机载激光雷达气溶胶后向散射系数反演的理论研究.  , 2011, 60(6): 064204. doi: 10.7498/aps.60.064204
    [13] 王敏, 胡顺星, 方欣, 汪少林, 曹开法, 赵培涛, 范广强, 王英俭. 激光雷达精确修正对流层目标定位误差.  , 2009, 58(7): 5091-5097. doi: 10.7498/aps.58.5091
    [14] 柯微娜, 程 茜, 钱梦騄. 测量单泡声致发光中气泡R(t)曲线的前向Mie散射技术.  , 2008, 57(6): 3629-3635. doi: 10.7498/aps.57.3629
    [15] 徐 涵, 常文蔚, 卓红斌. 短脉冲激光尾流场中的前向Raman散射.  , 2003, 52(1): 135-139. doi: 10.7498/aps.52.135
    [16] 吕亚军, 吴令安, 吴美娟, 李师群, 张培林. 光折变介质中前向三波混频的量子相关特性.  , 1998, 47(7): 1110-1117. doi: 10.7498/aps.47.1110
    [17] 何林生, 江海河, 夏宇兴. 由前向四波混频产生脉冲压缩态光场.  , 1991, 40(10): 1613-1623. doi: 10.7498/aps.40.1613
    [18] 徐至展, 唐永红, 钱爱娣. 激光等离子体受激布里渊散射光谱的周期性结构——前向散射的间接证据.  , 1988, 37(4): 557-565. doi: 10.7498/aps.37.557
    [19] 张富根. 圆偏振氟化氪激光在氢气中的前向受激喇曼散射.  , 1983, 32(9): 1211-1214. doi: 10.7498/aps.32.1211
    [20] 张才根, 张幼文. 环境辐射对目标热辐射特性测试的影响.  , 1981, 30(7): 953-961. doi: 10.7498/aps.30.953
计量
  • 文章访问数:  5037
  • PDF下载量:  127
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-02-05
  • 修回日期:  2021-07-21
  • 上网日期:  2021-08-15
  • 刊出日期:  2021-11-20

/

返回文章
返回
Baidu
map