基于深度学习的地空导弹发射区拟合算法

高晓光; 李新宇; 岳勐琪; 张金辉; 赵利强; 吴高峰; 李飞

doi:10.7527/S1000-6893.2019.22858

航空学报 >

2019 , Vol. 40 >Issue 9: 322858 - 322858

DOI: https://doi.org/10.7527/S1000-6893.2019.22858

电子电气工程与控制

基于深度学习的地空导弹发射区拟合算法

高晓光 ,
李新宇 ,
岳勐琪 ,
张金辉 ,
赵利强 ,
吴高峰 ,
李飞

展开

1. 西北工业大学电子信息学院, 西安 710072;
2. 航空工业洛阳电光设备研究所, 洛阳 471000;
3. 中国空空导弹研究院, 洛阳 471009

收稿日期: 2018-12-18

修回日期: 2019-02-22

网络出版日期: 2019-04-24

基金资助

国家自然科学基金（61573285）

收起

Fitting algorithm of ground-to-air missile launching area based on deep learning

GAO Xiaoguang ,
LI Xinyu ,
YUE Mengqi ,
ZHANG Jinhui ,
ZHAO Liqiang ,
WU Gaofeng ,
LI Fei

Expand

1. School of Electronics and Information, Northwestern Polytechnical University, Xi'an 710072, China;
2. AVIC Luoyang Institute of Electro-Optical Devices, Luoyang 471000, China;
3. China Airborne Missile Academy, Luoyang 471009, China

Received date: 2018-12-18

Revised date: 2019-02-22

Online published: 2019-04-24

Supported by

National Natural Science Foundation of China (61573285)

Fold

摘要

目前地空导弹发射区的拟合算法主要是多项式拟合法和BP神经网络拟合法。多项式拟合法存在函数形式难以确定、函数范围不易分段等问题，且拟合精度较低；传统神经网络方法要想达到较高精度，需要大量的隐层节点，且在隐层节点数增加到一定程度后，训练变得十分困难且精度很难继续提高。同时，传统神经网络需要大量的标签数据，进一步增大了实际应用的难度。为此，基于深度学习理论，设计了一种基于堆栈稀疏自编码器（SSAE）的深度拟合网络（DFN），并给出了相应的训练策略。仿真实验表明其相比传统算法具有更小的拟合误差优势。所设计的深度稀疏自编码器网络可以克服多项式拟合和传统神经网络的不足，不仅可以在大量无标签数据和少量标签数据条件下进行学习训练，而且可以进一步提升地空导弹发射区的拟合精度。

关键词： 地空导弹发射区; 神经网络; 深度学习; 堆栈稀疏自编码器; 深度拟合网络

本文引用格式

高晓光 , 李新宇 , 岳勐琪 , 张金辉 , 赵利强 , 吴高峰 , 李飞 . 基于深度学习的地空导弹发射区拟合算法[J]. 航空学报, 2019 , 40(9) : 322858 -322858 . DOI: 10.7527/S1000-6893.2019.22858

Abstract

At present, the fitting algorithm for ground-to-air missile launching areas mainly include the polynomial fitting algorithm and the Back Propagation (BP) neural network fitting method. The former is problematic in that the function form is difficult to determine, the function range is not easy to segment, and the fitting precision is low, whereas the latter requires a large number of hidden layer nodes to achieve higher precision. When the number of hidden layer nodes increases to a certain extent, its training becomes very difficult and its precision is difficult to keep improving. At the same time, traditional neural networks require a large amount of labeled data, which further increases the difficulty of practical applications. So, this paper designs a Stacked Sparse Auto-Encoder (SSAE) Deep Fitting Network (DFN) based on the deep learning theory, and provides the corresponding training strategy. The simulation results show that the proposed design has the advantage of getting lower fitting error compare to traditional methods. The deep sparse auto-encoder network designed in this paper can overcome the shortcomings of polynomial fitting and traditional neural network. This design can learn training with a large amount of unlabeled data and a small amount of tag data, further enhancing the ground-to-air missile launching area fitting accuracy.

Key words： ground-to-air missile launching area; neural network; deep learning; stack sparse auto-encoder; deep fitting network

参考文献

[1] 张振中, 鲁艺, 张辉明, 等. 地空导弹拦截模型下无人机突防方向的研究[J]. 计算机仿真, 2016, 33(9):113-131. ZHANG Z Z, LU Y, ZHANG H M, et al. Study on penetration direction of UAV based on the surface to air missile interception model[J]. Computer Simulation, 2016, 33(9):113-131(in Chinese).
[2] 李枭扬, 周德云, 冯琦, 等. 基于遗传规划的空空导弹攻击区拟合[J]. 弹箭与制导学报, 2015, 35(3):16-22. LI X Y, ZHOU D Y, FENG Q, et al. Air-to-air missile launch envelops fitting based on genetic programming[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2015, 35(3):16-22(in Chinese).
[3] 张振兴, 杨任农, 房育, 等. 基于改进BP神经网络的导弹攻击区计算[J]. 飞行力学, 2018, 36(2):48-52. ZHANG Z X, YANG R N, FANG Y, et al. Calculation of missile launch envelopes based on dynamic improved structured BP neural network[J]. Flight Dynamics, 2018, 36(2):48-52(in Chinese).
[4] 姚裕盛, 徐开俊. 基于BP神经网络的飞行训练品质评估[J]. 航空学报, 2017, 38(S1):721513. YAO Y S, XU K J. Quality assessment of flight training based on BP neural network[J]. Acta Aeronautica et Astronautica Sinica, 2017, 38(S1):721513(in Chinese).
[5] 周旺旺, 姚佩阳,张杰勇, 等. 基于深度神经网络的空中目标作战意图识别[J]. 航空学报, 2018,39(11):322468. ZHOU W W, YAO P Y, ZHANG J Y, et al. Combat intention recognition for aerial targets based on deep neural network[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(11):322468(in Chinese).
[6] 王正魁, 靳旭红,朱志斌, 等. 超声速湍流密度脉动预测的神经网络方法[J]. 航空学报,2018,39(10):122244. WANG Z K, JIN X H, ZHU Z B, et al. Neural network method for predicting density fluctuations in supersonic turbulence[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(10):122244(in Chinese).
[7] 李健伟, 曲长文, 彭书娟, 等. 基于卷积神经网络的SAR图像舰船目标检测[J]. 系统工程与电子技术, 2018, 40(9):1953-1959. LI J W, QU C W, PENG S J, et al. Ship detection in SAR images based on convolutional neural network[J]. Systems Engineering and Electronics, 2018, 40(9):1953-1959(in Chinese).
[8] 张天骐, 张婷, 熊梅, 等. 基于神经网络的低信噪比CBOC信号组合码序列盲估计[J]. 系统工程与电子技术, 2018, 40(12):2824-2832. ZHANG T Q, ZHANG T, XIONG M, et al. Neural network approach to blind-estimation of combined code sequence in lower SNR CBOC signals[J]. Systems Engineering and Electronics, 2018, 40(12):2824-2832(in Chinese).
[9] 周兴旺, 从福仲, 庞世春. 基于BN-and-BP神经网络融合的陆空联合作战效能评估[J]. 火力与指挥控制, 2018, 43(4):3-8. ZHOU X W, CONG F Z, PANG S C, et al. Effectiveness evaluation of air-ground joint operations based on the BN-and-BP neural network[J]. Fire Control & Command Control, 2018, 43(4):3-8(in Chinese).
[10] HINTON G E, SALAKHUTDINOV R R. Reducing the di mensionality of data with neural networks[J]. Science, 2006, 313(5786):504-507.
[11] 娄寿春. 地空导弹射击指挥控制模型[M]. 北京:国防工业出版社, 2009. LOU S C. Shooting command and control model of Ground-to-air missile[M]. Beijing:National Defense Industry Press, 2009(in Chinese).
[12] 谢世华, 肖毅于. 基于多项式拟合的GPS周跳探测研究[J]. 资源信息与工程, 2018, 33(5):135-138. XIE S H, XIAO Y Y. Research on GPS cycle slip detection based on polynomial fitting[J]. Resource Information and Engineering, 2018, 33(5):135-138(in Chinese).
[13] 李飞, 高晓光, 万开方. 基于动态Gibbs采样的RBM训练算法研究[J]. 自动化学报, 2016, 42(6):931-942. LI F, GAO X G, WAN K F. Research on RBM training algorithm with dynamic Gibbs sampling[J]. Acta Automatica Sinica, 2016, 42(6):931-942(in Chinese).
[14] GROSSBERG S, Nonlinear neural networks:Principles, mechanisms, and architectures[J]. Neural Networks, 1988, (1):15-57
[15] SERGEY I, CHRISTIAN S. Batch normalization:Accelerating deep network training by reducing internal covariate shift[C]//32nd International Conference on Machine Learning, 2015:448-456.
[16] DUMITRU E. Understanding deep architectures and the effect of unsupervised pre-training[D]. Quebec:University of Montréal, 2010.
[17] JAITLY N, NGUYEN P, SENIOR A, et al. Application of pretrained deep neural networks to large vocabulary conversational speech recognition[C]//Interspeech, 2012:2578-2581.
[18] POON H,DOMINGOS P. Sum-product networks:A new deep architecture[C]//2011 IEEE International Conference on Computer Vision Workshops. Piscataway, NJ:IEEE Press, 2011:689-690.
[19] DENG L. Design and learning of output representations for speech recognition[J]. Neural Information Processing Systems (NIPS) Workshop on Learning Output Representations, 2013, 35(2):156-162.
[20] VINCENT P, LAROCHELLE H, BENGIO Y, et al. Extracting and composing robust features with denoising autoencoders[C]//ICML 2008 Proceedings of the 25th International Conference on Machine Learning, 2008:1096-1103.
[21] SCHAUL T, ZHANG S, LECUN Y. No more pesky learning rates[J/OL]. ArXiv (2013-02-18)[2018-12-18]. https://arxiv.org/abs/1206.1106.
[22] LAN G. An optimal method for stochastic composite optimization[J]. Mathematical Programming, 2012, 133(1-2):365-397.
[23] YOSHUA B,OLIVIER D. On the expressive power of deep architectures[C]//Processing of the 14th International Conference on Discovery Science. Berlin:Springer-Verlag, 2011:18-36.
[24] YOSHUA B, PASCAL L, DAN P, et al. Greedy layer-wise training of deep networks[C]//Processing of the 12th Annual Conference on Neural Information Processing System, 2006:153-160.
[25] GLOROT X,YOSHUA B. Understanding the difficulty of training deep feedforward neural networks[C]//Processing of the 13th International Conference on Artifical Intelligence and Statistics, 2010:249-256.
[26] SALAKHUTDINOV R. Learning deep generative models[D]. Toronto:University of Toronto, 2009.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献