基于深度学习的空空导弹多类攻击区实时解算

doi:10.3969/j.issn.1000-1093.2020.12.012

摘要/Abstract

摘要： 现代空战日趋复杂，传统攻击区只能提供导弹发射的限度，无法满足现代空战决策的需求。基于目标机水平逃逸角度的最大攻击区、50°攻击区、90°攻击区、水平不可逃逸攻击区、最小攻击区等5类攻击区，针对现有攻击区解算方法无法同时解算多种攻击区的问题，提出多函数深度拟合网络(MFDFN)模型，以实现多种攻击区的同时解算。设计了改进的进退法解算流程，通过弹道仿真获取攻击区数据样本库。根据多函数拟合网络的特点，设计了“整体预训练+局部微调”训练策略，并对网络进行有监督训练。仿真结果表明：采用“整体预训练+局部微调”训练策略的MFDFN比传统网络不仅提高了计算实时性，而且很大程度上提高了计算准确性，其平均相对误差低至0.27%，平均绝对误差低至58.81 m；MFDFN模型具有较强的可靠性和实用性。

关键词: 空空导弹, 攻击区, 深度学习, 解算, 进退法, 多函数拟合

Abstract: Modern air combat is becoming more and more complex，and the traditional attack zones can only provide the limits of missile launch，which cannot meet the needs of modern air combat decision-making. For this reason，five types of attack zones, maximun attack zone, 50° attack zone, 90° attack zone, horizontal unescapable attack zone, and minimun attack zone, based on escape angles of enemy aircraft are proposed. The existing solving method of attack zone cannot be used to simultaneously solve the problem of multiple attack zones.A multi-function deep fitting network (MFDFN) is proposed to realize the simultaneous solution of multiple attack zones. Firstly，an improved advance-retreat method is designed，and the sample library of capture zone is obtained through ballistic simulation. According to the characteristics of multi-function fitting network，a training strategy，called “overall pre-training and local fine-tuning”，is presented，by which network is supervised trained.The simulated results show that the MFDFN using the “overall pre-training and local fine-tuning” training strategy reduces the computing time while greatly improving the computing accuracy. The average relative error is about 0.27%，and the average absolute error is about 58.81 meters，which proves that the model is reliable and practical.

Key words: air-to-airmissile, attackzone, deeplearning, computing, advance-retreatmethod, multi-functionfitting

中图分类号:

TJ762.2⁺3

闫孟达，杨任农，左家亮，胡东愿，岳龙飞，赵雨. 基于深度学习的空空导弹多类攻击区实时解算[J]. 兵工学报, 2020, 41(12): 2466-2477.

YAN Mengda， YANG Rennong， ZUO Jialiang， HU Dongyuan， YUE Longfei， ZHAO Yu. Real-time Computing of Air-to-air Missile Multiple Capture Zones Based on Deep Learning[J]. Acta Armamentarii, 2020, 41(12): 2466-2477.

参考文献

［1］吴文海，周思羽，高丽，等.基于导弹攻击区的超视距空战态势评估改进[J].系统工程与电子技术，2011，33(12): 2679-2685.
WU W H，ZHOU S Y，GAO L，et al.Improvements of situation assessment for beyond-visual-range air combat based on missile launching envelope analysis[J].Systems Engineering and Electronics，2011，33(12):2679-2685.(in Chinese)
[2］赵克新，黄长强，魏政磊，等.改进决策树的无人机空战态势估计[J].哈尔滨工业大学学报，2019，51(4): 66-73.
ZHAO K X，HUANG C Q，WEI Z L，et al. Situation assessment for unmanned aerial vehicle air combat based on anti-reasoning rules decision tree[J].Journal of Harbin Institute of Technology，2019，51(4):66-73.(in Chinese)
[3］ KUNG C C，CHIANG F L.A study of missile maximum capture area and fighter minimum evasive range for negotiation team air combat[C]∥Proceedings of the 15th International Conference on Control，Automation and Systems. Edinburgh，UK: IEEE，2015:207-212.
[4］左家亮，杨任农，张滢，等.基于启发式强化学习的空战机动智能决策[J].航空学报，2017，38(10):217-230.
ZUO J L，YANG R N，ZHANG Y，et al.Intelligent decision-making in air combat maneuvering based on heuristic reinforcement learning[J].Acta Aeronautica et Astronautica Sinica，2017，38(10): 217-230.(in Chinese)
[5］杨任农，张振兴，房育寰，等.深度置信网络在导弹攻击区分类中的应用[J].国防科技大学学报，2019，41(2): 98-106.
YANG R N，ZHANG Z X，FANG Y H，et al.Application of deep belief network in classification of missile launch envelopes[J].Journal of National University of Defense Technology，2019，41(2): 98-106.(in Chinese)
[6］张安柯，孔繁峨，贺建良.目标强机动对中远程空空导弹可攻击区的影响[J].弹箭与制导学报，2016，36(2): 21-25.
ZHANG A K，KONG F E，HE J L.Effect of strong maneuver target on attacking area of long-range air-to-air missile[J].Journal of Projectiles，Rockets，Missiles and Guidance，2016，36(2):21-25.(in Chinese)
[7］方学毅，刘俊贤，周德云.基于背景插值的空空导弹攻击区在线模拟方法[J].系统工程与电子技术，2019，41(6): 1286-1293.
FANG X Y，LIU J X，ZHOU D Y.Background interpolation for on-line simulation of capture zone[J].Systems Engineering and Electronics，2019，41(6):1286-1293.(in Chinese)
[8］惠耀洛，南英，陈哨东，等.空空导弹动态攻击区的高精度快速算法研究[J].弹道学报，2015，27(2):39-45.
HUI Y L，NAN Y，CHEN S D，et al. Research on rapid and high-precision calculation of dynamic attack zone for air-to-air missile[J].Journal of Ballistics，2015，27(2): 39-45.(in Chinese)
[9］ WU H X，QIU K，HUANG W，et al.Approximating solution for weapons employment zone of mid-range air-to-air missile[C]∥Proceedings of the 7th International Conference on Intelligent Human-machine Systems and Cybernetics.Hangzhou，China：IEEE，2015.
[10］ PARK S S，KIM D W，HONG J H，et al.Computation algorithm for dynamic launch zone of air-to-air missiles[J].Journal of the Korean Society for Aeronautical & Space Sciences，2014，42(9):762-772.
[11］ HUI Y L，NAN Y，CHEN X D，et al.Dynamic attack zone of air-to-air missile after being launched in random wind field[J].Chinese Journal of Aeronautics，2015，28(5): 1519-1528.
[12］史振庆，梁晓龙，张佳强，等.基于协同攻击区的航空集群最优空间构型研究[J].兵工学报，2019，40(4):788-798.
SHI Z Q，LIANG X L，ZHANG J Q，et al. Study of optimal spatial configuration of aircraft swarm based on cooperative attack zone[J].Acta Armamentarii，2019，40(4):788-798.(in Chinese)
[13］高晓光，李新宇，岳勐琪，等.基于深度学习的地空导弹发射区拟合算法研究[J/OL].航空学报，2019. (2019-05-31)[2019-11-06]. http:∥kns.cnki.net/kcms/detail/11.1929.V.20190531.1450.002.html.
GAO X G，LI X Y，YUE M Q，et al.Fitting algorithm of ground-to-air missile launching area based on deep learning [J/OL].Acta Aeronautica et Astronautica Sinica，2019. (2019-05-31)[2019-11-06].(in Chinese) http:∥kns.cnki.net/kcms/detail/11.1929.V.20190531.1450.002.html.(in Chinese)
[14］孟博.基于BP神经网络的空空导弹攻击大机动目标攻击区仿真研究[J].弹箭与制导学报，2017，37(4): 43-46，50.
MENG B.Research on launch envelopes simulation of air-to-air missile attacking high maneuvering targets based on BP neural network[J].Journal of Projectiles，Rockets，Missiles and Guidance， 2017，37(4):43-46，50.(in Chinese)
[15］张振兴，杨任农，房育寰，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 H，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)
[16］ SCHARL J，MAVRIS D，BURDUN I.Use of flight simulation in early design-formulation and application of the virtual testing and evaluation methodology[C]∥Proceedings of 2000 World Aviation Conference.San Diego，CA，US: AIAA，2000:5590.
[17］ ALKAHER D，MOSHAIOV A.Dynamic-escape-zone to avoid energy- bleeding coasting missile[J].Journal of Guidance，Control，and Dynamics，2015，38(10) 1908-1921.
[18］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):322468-322476.
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):322468-322476.(in Chinese)
[19］ GLOROT X，BORDES A，BENGIO Y.Deep sparse rectifier neural networks[C]∥Proceedings of the 14th International Conference on Artificial Intelligence and Statistics. Fort Lauderdale，FL，US:Society for Artificial Intelligence and Statistics，2011: 315-323.
[20］ KLAMBAUER G，UNTERTHINER T，MAYR A，et al.Self-normalizing neural networks[C]∥Proceedings of the 31st Advances in Neural Information Processing Systems. Long Beach，CA，US: Neural Information Processing Systems，2017:971-980.
[21］ SERGEY I，CHRISTIAN S.Batch normalization: accelerating deep network training by reducing internal covariate shift[C]∥Proceedings of the 32nd International Conference on Machine Learning.New York，NY，US:International Machine Learning Society，2015：448-456.
[22］ HINTON G E，SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks[J].Science，2006，313(5786): 504-507.
[23］ KINGMA D P，BA J.Adam:a method for stochastic optimization[EB/OL].San Diego，CA，US:ICLR，2015.(2017-01-30)[2019-11-06].https:∥arxiv.org/abs/1412.6980.

[1]	田大明, 苗圃. 融合模型求解与深度学习的可见光通信非线性均衡器[J]. 兵工学报, 2024, 45(2): 466-473.
[2]	张堃, 杜睿怡, 时昊天, 华帅. 基于Mogrifier-BiGRU的飞行器轨迹预测[J]. 兵工学报, 2024, 45(2): 373-384.
[3]	杨家铭, 潘悦, 王强, 曹怀刚, 高荪培. 水下弱目标跟踪的深度学习方法研究[J]. 兵工学报, 2024, 45(2): 385-394.
[4]	申程, 张连超, 张卓, 朱文亮, 陈雨康. 轻量化弹道解算系统火控修正和瞄准线滤波预测[J]. 兵工学报, 2024, 45(2): 429-442.
[5]	秦昊林, 许廷发, 李佳男. 基于超像素注意力和孪生结构的半监督高光谱显著性目标检测[J]. 兵工学报, 2023, 44(9): 2639-2649.
[6]	刘畅, 雷红波, 林时尧, 范世鹏, 王江. 基于多模型网络的激光末制导炮弹诸元解算方法[J]. 兵工学报, 2023, 44(9): 2745-2755.
[7]	彭沛然, 任术波, 李佳男, 周鸿伟, 许廷发. 基于光照感知的多光谱融合行人检测方法[J]. 兵工学报, 2023, 44(9): 2622-2630.
[8]	周宇, 曹荣刚, 栗苹, 马啸. 一种用于外场试验图像的引信炸点检测方法[J]. 兵工学报, 2023, 44(8): 2453-2464.
[9]	张良安, 陈洋, 谢胜龙, 刘同鑫. 基于机器视觉与深度学习的飞机防护栅裂纹检测系统[J]. 兵工学报, 2023, 44(2): 507-516.
[10]	王洋, 冯永新, 宋碧雪, 田秉禾. DP-DRCnet卷积神经网络信号调制识别算法[J]. 兵工学报, 2023, 44(2): 545-555.
[11]	郑志伟, 管雪元, 傅健, 马训穷, 尹上. 基于卷积神经网络与长短期记忆神经网络的弹丸轨迹预测[J]. 兵工学报, 2023, 44(10): 2975-2983.
[12]	王强, 吴乐天, 李红, 王勇, 王欢, 杨万扣. 基于双支网络协作的红外弱小目标检测[J]. 兵工学报, 2023, 44(10): 3165-3176.
[13]	王亮，陈建华，李烨. 一种基于深度学习的无人艇海上目标识别技术[J]. 兵工学报, 2022, 43(S2): 13-19.
[14]	王晓琦，赵旸，张键，王硕. 基于深度学习的船体三维模型自动生成方法[J]. 兵工学报, 2022, 43(S2): 115-119.
[15]	李良福，陈卫东，高强，许开銮，刘轩，何曦，钱钧. 基于深度学习的光电系统智能目标识别[J]. 兵工学报, 2022, 43(S1): 162-168.