基于鱼侧线感知原理和深度学习的水下平动目标方向识别

doi:10.3901/JME.2020.12.231

机械工程学报 ›› 2020, Vol. 56 ›› Issue (12): 231-239.doi: 10.3901/JME.2020.12.231

扫码分享

基于鱼侧线感知原理和深度学习的水下平动目标方向识别

张勇^1,2, 郑贤德^1,2, 季明江^1,2, 林鑫^1,2, 邱静^1,2, 刘冠军^1,2

1. 国防科技大学智能科学学院长沙 410073;
2. 国防科技大学装备综合保障技术重点实验室长沙 410073

收稿日期:2019-12-28 修回日期:2020-05-09 出版日期:2020-06-20 发布日期:2020-07-14
通讯作者: 刘冠军(通信作者),男,1972年出生,博士,研究员,博士研究生导师。主要研究方向为装备智能状态感知、仿生感知及新型传感等。E-mail:gjliu342@qq.com
作者简介:张勇,男,1982年出生,博士。主要研究方向为装备智能状态感知等。E-mail:zhangyong@nudt.edu.cn
基金资助:
国家自然科学基金资助项目（51675528，51605482）。

Underwater Translational Target Direction Recognition Based on Lateral Line Perception Principle and Deep Learning

ZHANG Yong^1,2, ZHENG Xiande^1,2, JI Mingjiang^1,2, LIN Xin^1,2, QIU Jing^1,2, LIU Guanjun^1,2

1. College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073;
2. Science and Technology on Integrated Logistics Support Laboratory, National University of Defense Technology, Changsha 410073

Received:2019-12-28 Revised:2020-05-09 Online:2020-06-20 Published:2020-07-14

摘要/Abstract

摘要： 鱼侧线感知原理为潜航器水下目标感知技术研究提供了一种新的思路，但由于流场模型难以准确构建，基于模型的解析方法难以准确感知水下平动目标，提出基于鱼侧线感知原理和深度学习的水下平动目标方向识别方法。通过构建偶极子源周围压力场分布模型分析鱼侧线感知目标原理，在理论上分析压力变化与偶极子源尺寸、运动参数、位置的关系，理论分析表明，流场中压力的大小与偶极子源的位置密切相关，偶极子源振动频率处的压力变化特征明显，可用于训练和识别。采用时频分析方法处理压力传感器信号并提取时频分布特征，研究表明不同平动方向产生的压力变化具有不同的时频分布特征。提出利用卷积神经网络训练压力传感器信号，进而识别水下平动目标方向。在十字形传感器阵列及试验平台上开展试验验证，试验结果表明水下平动目标方向综合识别准确率在80%以上。在无需准确建立流场模型的情况下，通过深度学习可较准确识别水下平动目标，为潜航器水下目标感知提供了一种新的技术途径。

关键词: 仿生鱼侧线, 深度学习, 水下平动目标方向识别

Abstract: The fish lateral line perception principle provides a new idea for the submarine underwater target sensing technology. Because the flow field model is difficult to build accurately, the model-based analytical method is difficult to precisely detect the target. The underwater translational target direction recognition method is proposed based on lateral line perception principle and deep learning. By constructing a pressure field distribution model around the dipole source, the lateral line perception principle is analyzed, and the relationship between pressure changes and the dipole source size, motion parameters, and position is theoretically analyzed. The results show that the pressure in the flow field is related to the position of the dipole source, and the features of pressure change at the dipole source vibration frequency are obvious, which can be used for training and identification. The time-frequency analysis method is used to process the pressure sensor signals and extract the time-frequency distribution features. Studies show that pressure changes in different translational directions have different time-frequency distribution features. It is proposed to use the convolutional neural networks to train the pressure sensor signals to identify the direction of the underwater translation target. The cross-shaped sensor array and the test platform are used to carry out test. Results show that the accuracy of the comprehensive recognition is above 80%. Without the need to accurately establish a flow field model, underwater translation targets can be accurately identified through deep learning. A new technical approach is provided for submarine underwater target perception.

Key words: bionic lateral line, deep learning, underwater translation direction recognition

中图分类号:

TP391
TJ67

张勇, 郑贤德, 季明江, 林鑫, 邱静, 刘冠军. 基于鱼侧线感知原理和深度学习的水下平动目标方向识别[J]. 机械工程学报, 2020, 56(12): 231-239.

ZHANG Yong, ZHENG Xiande, JI Mingjiang, LIN Xin, QIU Jing, LIU Guanjun. Underwater Translational Target Direction Recognition Based on Lateral Line Perception Principle and Deep Learning[J]. Journal of Mechanical Engineering, 2020, 56(12): 231-239.

参考文献

[1] COOMBS S. Smart skins:Information processing by lateral line flow sensors[J]. Autonomous Robots,2001,11(3):255-261.
[2] COOMBS S,PETER G,HEINRICH M. The mecha-nosensory lateral line[M]. New York:Springer,1989.
[3] BECKMANN M,EROS T,SCHMITZ A,et al. Number and distribution of superficial neuromasts in twelve common European cypriniform fishes and their relation-ship to habitat occurrence[J]. International Review of Hydrobiology,2013,95:273-284.
[4] NETTEN S M V. Hydrodynamic detection by cupulae in a lateral line canal:Functional relations between physics and physiology[J]. Biological Cybernetics,2006,94:67-85.
[5] KOTTAPALLI A G P,ASADNIA M,MIAO J M,et al. Touch at a distance sensing:Lateral-line inspired MEMS flow sensors[J]. Bioinspiration & Biomimetics,2014,9(4):14.
[6] ABDULSADDA A T,TAN X B. Nonlinear estima-tion-based dipole source localization for artificial lateral line systems[J]. Bioinspiration & Biomimetics,2013,8(2):15.
[7] ABDULSADDA A T,TAN X B. Artificial lateral line-based localization of a dipole source with unknown vibration amplitude and direction[C]//IEEE 15th Inte-rnational Conference on Advanced Robotics:New Boun-daries for Robotics,2011,447-452.
[8] DAGAMSEH A,WIEGERINK R,LAMMERINK T,et al. Imaging dipole flow sources using an artificial lateral-line system made of biomimetic hair flow sens-ors[J]. Journal of the Royal Society Interface,2013,10(83):9.
[9] 刘贵杰,宫华耀,吴乃龙,等. 基于鱼类侧线感知机理的流场辨识方法及仿真研究[J]. 机械工程学报,2016,52(17):162-167. LIU Guijie,GONG Huayao,WU Nailong,et al. Simulation research in water condition recognition method based on fish lateral line sensing mechanism[J]. Journal of Mechanical Engineering,2016,52(17):162-167.
[10] 吴乃龙,吴超,葛彤,等. 基于鱼类体线感知机理的水下机器人水流场识别研究[J]. 机械工程学报,2016,52(13):54-59. WU Nailong,WU Chao,GE Tong,et al. Flow recognition of underwater vehicle based on the perception mechanism of lateral line[J]. Journal of Mechanical Engineering,2016,52(13):54-59.
[11] ZHENG X,WANG C,FAN R,et al. Artificial lateral line based local sensing between two adjacent robotic fish[J]. Bioinspiration & Biomimetics,2017,13(1):016002.
[12] PERERA T B D. Fish can encode order in their spatial map[J]. Proceedings of the Royal Society of London. Series B:Biological Sciences,2004,271(1553):2131-2134.
[13] MONTGOMERY J C,Macdonald J A. Sensory tuning of lateral line receptors in antarctic fish to the movements of planktonic prey[J]. Science,1987,235(4785):195-196.
[14] SCHWALBE M A,BASSETT D K,WEBB J F. Feeding in the dark:Lateral-line-mediated prey detection in the peacock cichlid Aulonocara stuartgranti[J]. Journal of Experimental Biology,2012,215(12):2060-2071.
[15] BAKER C F,MONTGOMERY J C. The lateral line can mediate rheotaxis in fish[J]. Nature,1997,389(6654):960-963.
[16] STOKES G G. On the effect of the internal friction of fluids on the motion of pendulums[J]. Transactions of the Cambridge Philosophical Society,1851,9:8.
[17] GANLEY T,HUNG D L S,ZHU G,et al. Modeling and inverse compensation of temperature-dependent ionic polymer-metal composite sensor dynamics[J]. IEEE/ASME Transactions on Mechatronics,2011,16:80-89.
[18] ABDULSADDA A T,TAN X B. An artificial lateral line system using IPMC sensor arrays[J]. International Jou-rnal of Smart & Nano Materials,2012,3:226-242.
[19] ZHENG X D,ZHANG Y,JI M J,et al. Underwater positioning based on an artificial lateral line and a generalized regression neural network[J]. Journal of Bionic Engineering,2018,15:883-893.
[20] JI M J,ZHANG Y,ZHENG X D,et al. A fish-shaped minimal prototype of lateral line system based on pressure sensing[C]//IEEE International Conference on Mechatr-onics & Automation,Takamatsu,Japan,2017,596-601.
[21] JI M,ZHANG Y,ZHENG X,et al. Resolution impro-vement of dipole source localization for artificial lateral lines based on multiple signal classification[J]. Bioinspiration & Biomimetics,2019,14(1):016016.
[22] SRIVASTAVA N,HINTON G,KRIZHEVSKY A,et al. Dropout:A simple way to prevent neural networks from overfitting[J]. Journal of Machine Learning Research,2014,15(1):1929-1958.
[23] HINTON G E,OSINDERO S,TEH Y W. A fast learning algorithm for deep belief nets[M]. Massachusetts:MIT Press,2006.
[24] KRIZHEVSKY A,SUTSKEVER I,HINTON G E. ImageNet classification with deep convolutional neural networks[C]//International Conference on Neural Infor-mation Processing Systems. Curran Associates Inc. 2012:1097-1105.
[25] ZEILER M D,FERGUS R. Visualizing and underst-anding convolutional networks[C]//European Conference on Computer Vision. Springer,Cham,2014:818-833.

基于鱼侧线感知原理和深度学习的水下平动目标方向识别

Underwater Translational Target Direction Recognition Based on Lateral Line Perception Principle and Deep Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	李春凯, 王嘉昕, 石玗, 代悦. GTAW熔池激光条纹动态行为与熔透预测方法研究[J]. 机械工程学报, 2024, 60(6): 236-244.
[2]	赵德望, 姜超, 赵延广, 杨文平, 樊俊铃. 铝锂合金疲劳性能及预测研究-基于试验和“浅层”+“深度”混合神经网络方法[J]. 机械工程学报, 2024, 60(16): 190-199.
[3]	付玲, 佘玲娟, 颜镀镭, 张鹏, 龙湘云. 基于内嵌物理信息与注意力机制BiLSTM神经网络的臂架系统疲劳损伤预测模型[J]. 机械工程学报, 2024, 60(13): 205-215.
[4]	陈钱, 陈康康, 董兴建, 皇甫一樊, 彭志科, 孟光. 一种面向机械设备故障诊断的可解释卷积神经网络[J]. 机械工程学报, 2024, 60(12): 65-76.
[5]	刘岳开, 王天杨, 褚福磊. 基于低延迟可解释性深度学习的复杂旋转机械关键部件知识嵌入与诊断方法研究[J]. 机械工程学报, 2024, 60(12): 107-115.
[6]	赖旭伟, 丁昆, 张楷, 黄锋飞, 郑庆, 李致萱, 丁国富. 基于可解释物理引导空间注意力改进的跨工艺参数立铣刀磨损辨识[J]. 机械工程学报, 2024, 60(12): 147-157.
[7]	林昕, 毛杨坤, 傅盈西, 朱锟鹏. 基于图像语义分割与图卷积的选区激光熔融成形过程羽流运动特征分析[J]. 机械工程学报, 2024, 60(12): 168-182.
[8]	邵海东, 肖一鸣, 邓乾旺, 任颖莹, 韩特. 基于不确定性感知网络的可信机械故障诊断[J]. 机械工程学报, 2024, 60(12): 194-206.
[9]	李玖法, 邹博文, 任玥. 考虑车-路交互作用的车辆轨迹预测算法研究[J]. 机械工程学报, 2024, 60(10): 76-85.
[10]	许思远, 张卫东, 毛玉明, 牛斌. 数据驱动的指定变形非均质多胞结构优化设计[J]. 机械工程学报, 2024, 60(1): 85-95.
[11]	何赟泽, 李响, 王洪金, 侯岳骏, 张帆, 牟欣颖, 刘浩, 程豪, 李时华, 李杰. 基于可见光和热成像的风机叶片全周期无损检测综述[J]. 机械工程学报, 2023, 59(6): 32-45.
[12]	康杰, 王寅, 罗杰, 孙嘉宝, 曾舒洪. 自主工作模态分析方法研究综述[J]. 机械工程学报, 2023, 59(13): 89-109.
[13]	吴强威, 文龙. 基于联合对抗训练的深度学习表面缺陷检测方法[J]. 机械工程学报, 2023, 59(12): 173-182.
[14]	李雄, 苏建宁, 张志鹏. 基于深度学习的产品概念草图生成设计研究[J]. 机械工程学报, 2023, 59(11): 16-30.
[15]	段暕, 周宏娣, 刘智勇, 詹小斌, 梁健强, 史铁林. 基于改进PCANet模型的铣刀磨损预测方法研究[J]. 机械工程学报, 2023, 59(1): 278-285.