代价敏感的空中目标意图识别方法

doi:10.7527/S1000-6893.2023.28551

Abstract

Abstract:

Air intention recognition is the key to the transition from the information domain to the cognitive domain， serving as the basis for command and decision making in air defense operations， as well as the prerequisite and foundation for battlefield cognition and intelligent decision making. It has always been considered the core content of battlefield situational awareness. However， existing research mostly focuses solely on the accuracy of intention recognition， without considering whether the misjudgment costs of intention recognition are equivalent. Air target intention recognition should consider both accuracy and cost sensitivity. To solve the problem of different misjudgments that may cause different losses to our side， a new deep learning model （GRU-FCN） for cost sensitive aerial target intention recognition is designed based on gated cyclic units and fully convolutional networks， and a cost sensitive improvement strategy is also proposed. Experimental analysis shows that the accuracy of the GRU-FCN model reaches 98.57%， surpassing other aerial target intention recognition models in the comparative experiment by 2.24%. After incorporating the cost sensitive improvement strategy， the overall misjudgment cost has been reduced from 0.346 7 to 0.175 7， ensuring that the accuracy meets the requirements while also possessing the ability to recognize cost sensitive intentions.

Key words: deep learning, cost sensitive, aerial target, intention recognition, Gated Recurrent Unit (GRU), Fully Convolutional Networks (FCN)

CLC Number:

V219

Peng DING, Yafei SONG. A cost-sensitive method for aerial target intention recognition[J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(24): 328551-328551.

Figures/Tables 24

Fig.1

Fig.2

Table 1

Fig.3

Fig.4

Fig. 5

Fig.6

Table 2

Table 3

Table 4

Table 5

Fig.7

Fig.8

Fig.9

Table 6

Table 7

Fig.10

Fig.11

Table 8

Fig.12

Table 9

Table 10

Fig.13

Table 11

References 38

1	姚庆锴，柳少军，贺筱媛，等. 战场目标作战意图识别问题研究与展望［J］. 指挥与控制学报， 2017， 3（2）： 127-131.
	YAO Q K， LIU S J， HE X Y， et al. Research and prospect of battlefield target operational intention recognition［J］. Journal of Command and Control， 2017， 3（2）： 127-131 （in Chinese）.
2	XIA P C， CHEN M， ZOU J， et al. Prediction of air target intention utilizing incomplete information［C］∥Chinese Intelligent Systems Conference. Singapore： Springer， 2016： 395-403.
3	陈优敏，李长云. 基于知识图谱的目标战术意图识别仿真［J］. 计算机仿真， 2019， 36（8）： 1-4， 19.
	CHEN Y M， LI C Y. Simulation of target tactical intention recognition based on knowledge map［J］. Computer Simulation， 2019， 36（8）： 1-4， 19 （in Chinese）.
4	尹翔，张萌，陈梦乔. 基于判别分析的空中目标作战意图识别［J］. 弹箭与制导学报， 2018， 38（3）： 46-50.
	YIN X， ZHANG M， CHEN M Q. Combat intention recognition of the target in the air based on discriminant analysis［J］. Journal of Projectiles， Rockets， Missiles and Guidance， 2018， 38（3）： 46-50 （in Chinese）.
5	XU Y H， CHENG S Y， ZHANG H B， et al. Air target combat intention identification based on IE-DSBN［C］∥2020 International Workshop on Electronic Communication and Artificial Intelligence （IWECAI）. Piscataway： IEEE Press， 2020： 36-40.
6	XU X M， YANG R N， FU Y. Situation assessment for air combat based on novel semi-supervised naive Bayes［J］. Journal of Systems Engineering and Electronics， 2018， 29（4）： 768-779.
7	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）： 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）： 322476 （in Chinese）.
8	翟翔宇，杨风暴，吉琳娜，等. 标准化全连接残差网络空战目标威胁评估［J］. 火力与指挥控制， 2020， 45（6）： 39-44.
	ZHAI X Y， YANG F B， JI L N， et al. Air combat targets threat assessment based on standardized fully connected network and residual network［J］. Fire Control & Command Control， 2020， 45（6）： 39-44 （in Chinese）.
9	钱钊，刘钦，鹿瑶，等. 基于长短时神经网络的目标意图识别［J］. 太赫兹科学与电子信息学报， 2022， 20（11）： 1156-1162.
	QIAN Z， LIU Q， LU Y， et al. Identification of target’s combat intention based on long short term memory network［J］. Journal of Terahertz Science and Electronic Information Technology， 2022， 20（11）： 1156-1162 （in Chinese）.
10	胡智勇，刘华丽，龚淑君，等. 基于随机森林的目标意图识别［J］. 现代电子技术， 2022， 45（19）： 1-8.
	HU Z Y， LIU H L， GONG S J， et al. Target intention recognition based on random forest［J］. Modern Electronics Technique， 2022， 45（19）： 1-8 （in Chinese）.
11	TENG F， SONG Y F， GUO X P. Attention-TCN-BiGRU： An air target combat intention recognition model［J］. Mathematics， 2021， 9（19）： 2412.
12	WANG S Y， WANG G， FU Q， et al. STABC-IR： An air target intention recognition method based on bidirectional gated recurrent unit and conditional random field with space-time attention mechanism［J］. Chinese Journal of Aeronautics， 2023， 36（3）： 316-334.
13	TENG F， GUO X P， SONG Y F， et al. An air target tactical intention recognition model based on bidirectional GRU with attention mechanism［J］. IEEE Access， 2021， 9： 169122-169134.
14	QU C X， GUO Z C， XIA S J， et al. Intention recognition of aerial target based on deep learning［J］. Evolutionary Intelligence， 2022： 1-9.
15	BHATT D， PATEL C， TALSANIA H， et al. CNN variants for computer vision： History， architecture， application， challenges and future scope［J］. Electronics， 2021， 10（20）： 2470.
16	SHERSTINSKY A. Fundamentals of recurrent neural network （RNN） and long short-term memory （LSTM） network［J］. Physica D： Nonlinear Phenomena， 2020， 404： 132306.
17	LIPTON Z C， BERKOWITZ J， ELKAN C. A critical review of recurrent neural networks for sequence learning［DB/OL］. arXiv preprint：1506.00019， 2015.
18	LI X C， MA X F， XIAO F C， et al. Time-series production forecasting method based on the integration of Bidirectional Gated Recurrent Unit （Bi-GRU） network and Sparrow Search Algorithm （SSA）［J］. Journal of Petroleum Science and Engineering， 2022， 208： 109309.
19	DEY R， SALEM F M. Gate-variants of Gated Recurrent Unit （GRU） neural networks［C］∥2017 IEEE 60th International Midwest Symposium on Circuits and Systems （MWSCAS）. Piscataway： IEEE Press， 2017： 1597-1600.
20	LINDEMANN B， MÜLLER T， VIETZ H， et al. A survey on long short-term memory networks for time series prediction［J］. Procedia CIRP， 2021， 99： 650-655.
21	CHENG J P， DONG L， LAPATA M. Long short-term memory-networks for machine reading［DB/OL］. arXiv preprint： 1601.06733， 2016.
22	GRAVES A. Long short-term memory［M］∥Supervised Sequence Labelling with Recurrent Neural Networks. Berlin： Springer， 2012： 37-45.
23	CHENG Y， WANG D， ZHOU P， et al. Model compression and acceleration for deep neural networks： The principles， progress， and challenges［J］. IEEE Signal Processing Magazine， 2018， 35（1）： 126-136.
24	LA ROSA L E C， SOTHE C， FEITOSA R Q， et al. Multi-task fully convolutional network for tree species mapping in dense forests using small training hyperspectral data［J］. ISPRS Journal of Photogrammetry and Remote Sensing， 2021， 179： 35-49.
25	ELSAYED N， MAIDA A S， BAYOUMI M. Deep gated recurrent and convolutional network hybrid model for univariate time series classification［DB/OL］. arXiv preprint：1812.07683， 2018.
26	KARIM F， MAJUMDAR S， DARABI H， et al. LSTM fully convolutional networks for time series classification［J］. IEEE Access， 2017， 6： 1662-1669.
27	陈浩，任卿龙，滑艺，等. 基于模糊神经网络的海面目标战术意图识别［J］. 系统工程与电子技术， 2016， 38（8）： 1847-1853.
	CHEN H， REN Q L， HUA Y， et al. Fuzzy neural network based tactical intention recognition for sea targets［J］. Systems Engineering and Electronics， 2016， 38（8）： 1847-1853 （in Chinese）.
28	朱可钦，董彦非. 空战机动动作库设计方式研究［J］. 航空计算技术， 2001， 31（4）： 50-52.
	ZHU K Q， DONG Y F. Study on the design of air combat maneuver library［J］. Aeronautical Computer Technique， 2001， 31（4）： 50-52 （in Chinese）.
29	孙艺聪，田润澜，王晓峰，等. 基于改进CLDNN的辐射源信号识别［J］. 系统工程与电子技术， 2021， 43（1）： 42-47.
	SUN Y C， TIAN R L， WANG X F， et al. Emitter signal recognition based on improved CLDNN［J］. Systems Engineering and Electronics， 2021， 43（1）： 42-47 （in Chinese）.
30	FAWAZ H I， FORESTIER G， WEBER J， et al. Deep learning for time series classification： A review［J］. Data Mining and Knowledge Discovery， 2019， 33（4）： 917-963.
31	ZHAO B D， LU H Z， CHEN S F， et al. Convolutional neural networks for time series classification［J］. Journal of Systems Engineering and Electronics， 2017， 28（1）： 162-169.
32	RUIZ A P， FLYNN M， LARGE J， et al. The great multivariate time series classification bake off： A review and experimental evaluation of recent algorithmic advances［J］. Data Mining and Knowledge Discovery， 2021， 35（2）： 401-449.
33	PINNERI C， SAWANT S， BLAES S， et al. Sample-efficient cross-entropy method for real-time planning［DB/OL］. arXiv preprint： 2008.06389， 2020.
34	KIEFER J， WOLFOWITZ J. Stochastic estimation of the maximum of a regression function［J］. The Annals of Mathematical Statistics， 1952， 23（3）： 462-466.
35	TIELEMAN T， HINTON G. Lecture 6.5-rmsprop： Divide the gradient by a running average of its recent magnitude［J］. COURSERA： Neural Networks for Machine Learning， 2012， 4（2）： 26-31.
36	KINGMA D P， BA J. Adam： A method for stochastic optimization［DB/OL］. arXiv Preprint： 1412.6980， 2014.
37	滕飞，刘曙，宋亚飞. BiLSTM-Attention：一种空中目标战术意图识别模型［J］. 航空兵器， 2021， 28（5）： 24-32.
	TENG F， LIU S， SONG Y F. BiLSTM-attention： An air target tactical intention recognition model［J］. Aero Weaponry， 2021， 28（5）： 24-32 （in Chinese）.
38	ZHU W B， WEBB Z T， MAO K T， et al. A deep learning approach for process data visualization using t-distributed stochastic neighbor embedding［J］. Industrial & Engineering Chemistry Research， 2019， 58（22）： 9564-9575.

意图特征	特征描述	数据类型
高度	敌空中目标的海拔高度	数值型
敌我距离	敌空中目标与我军事中心的距离	数值型
速度	敌空中目标的飞行速度	数值型
加速度	敌空中目标的飞行加速度	数值型
航向角	敌空中目标飞行方向（正北方向为0°，顺时针方向一周分为360°）	数值型
方位角	我方军事建筑到敌空中目标方向的方位角（正北方向为0°，顺时针方向一周分为360°）	数值型
雷达反射面积	敌空中目标在我方雷达上的回波大小	数值型
敌机类型	敌战斗机的种类型号	非数值型
对空雷达状态	敌机的对空雷达是否开启	非数值型
对地雷达状态	敌机的对地雷达是否开启	非数值型
机动类型	敌机主要采取的机动方式	非数值型
干扰状态	敌机的干扰装置是否开启	非数值型

优化器	准确率/%
Adam	98.57
Adamax	96.52
RMSprop	97.10
SGD	94.81

学习率	准确率/%
0.001	98.57
0.002	97.86
0.004	97.19
0.006	96.28

超参数	数值
优化器（Optimizer）	Adam
批大小（Batch_size）	32
训练周期（Epoch）	200
学习率（Learning rate）	0.001
GRU层数（GRU_layers）	1
FCN 通道数（FCN out_channels）	128，256，128
FCN卷积核大小（FCN kernel_size）	8，5，3

参数	CAIR	CAIR-key
K₁	1.0	1.0
K₂	5.0	12.5
J₁	1.0	1.0
J₂	10.0	10.0

A cost-sensitive method for aerial target intention recognition

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 24

References 38

Related Articles 15

Recommended Articles

Metrics

Comments

模型	Accuracy/%	Cost	H-error	L-error
GRU-FCN	98.57	0.346 7	18	12
CAIR	96.14	0.175 7	4	77
CAIR-key	92.38	0.225 7	0	160

模型组成结构			Accuracy/%	Cost	H-error	L-error
GRU-FCN	C-CE	C-ACC	Accuracy/%	Cost	H-error	L-error
√			98.57	0.346 7	18	12
√	√		96.48	0.264 7	12	62
√		√	95.76	0.227 6	7	82
√	√	√	96.14	0.175 7	4	77

信噪比/dB	GRU-FCN		CAIR
信噪比/dB	准确率/%	Cost	准确率/%	Cost
无噪声	98.57	0.346 7	96.14	0.175 7
10	96.52	0.587 1	93.90	0.395 2
5	93.57	1.192 8	91.71	0.694 8
0	92.10	1.450 5	89.00	0.800 9
-5	91.19	1.663 5	87.48	0.935 7
-10	89.52	1.914 3	85.57	1.300 5

模型	Accuracy/%	Cost	H-error	L-error
LSTM	85.86	2.131 9	139	158
LSTM-CAIR	86.00	0.506 7	31	263
BiGRU-Attention	91.43	1.135 7	75	105
BiGRU-Attention-CAIR	90.14	0.411 0	29	178
Attention-TCN-BiGRU	96.33	0.709 5	36	41
Attention-TCN-BiGRU-CAIR	95.19	0.207 6	12	89
GRU-FCN	98.57	0.346 7	18	12
GRU-FCN-CAIR	96.14	0.175 7	4	77

意图类型	Accuracy/%
意图类型	①	②	③	④	⑤	⑥	⑦	⑧
攻击	77.0	88.7	91.3	95.3	95.0	91.7	99.7	99.3
突防	86.0	96.0	99.0	99.7	87.7	95.3	97.7	96.3
干扰	85.3	93.7	99.3	99.7	83.7	88.3	96.3	97.7
监视	88.3	91.0	97.7	99.0	85.7	94.0	96.7	98.0
侦察	85.0	92.3	94.0	99.3	76.7	84.3	90.3	94.3
佯攻	79.3	81.7	93.0	97.0	73.3	79.3	89.0	87.3
撤退	100	100	100	100	100	98.0	99.7	100

[1]	Pengyu LIU, Xueyao ZHU. Semantic parsing technology of air traffic control instruction in fusion airspace based on deep learning [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(S1): 727592-727592.
[2]	Lei HE, Weiqi QIAN, Kangsheng DONG, Xian YI, Congcong CHAI. Aerodynamic characteristics modeling of iced airfoil based on convolution neural networks [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(5): 126434-126434.
[3]	Xiaohang LI, Jianjiang ZHOU. Multi⁃scale modality fusion network based on adaptive memory length [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(22): 628977-628977.
[4]	Haowen LUO, Shaoming HE, Tianyu JIN, Zichao LIU. Impact-angle-constrained with time-minimum guidance algorithm based on transfer learning [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(19): 328400-328400.
[5]	Yunhe ZHAO, Shengnan WANG. Solution to stress intensity factor by weight function method based on deep learning [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(19): 228367-228367.
[6]	Shuyi GAO, Defu LIN, Duo ZHENG, Xinyu HU. Intelligent cooperative interception strategy of aircraft against cluster attack [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(18): 328301-328301.
[7]	Yubin YUAN, Yiquan WU, Langyue ZHAO, Jinlin CHEN, Qichang ZHAO. Research progress of UAV aerial video multi⁃object detection and tracking based on deep learning [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(18): 28334-028334.
[8]	Rongsheng ZHANG, Yansheng WU, Xudong QIN, Puzhuo ZHANG. A real⁃time in⁃flight wind estimation and prediction method based on deep learning [J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(13): 327860-327860.
[9]	Qiang WANG, Letian WU, Yong WANG, Huan WANG, Wankou YANG. An infrared small target detection method based on key point [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(10): 328173-328173.
[10]	SU Lingfei, HUA Yongzhao, DONG Xiwang, REN Zhang. Human-UAV swarm multi-modal intelligent interaction methods [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(S1): 727001-727001.
[11]	ZHAO Liangyu, LI Dan, ZHAO Chenyue, JIANG Fei. Some achievements on detection methods of UAV autonomous landing markers [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(9): 25882-025882.
[12]	CAO Ming, WANG Peng, ZUO Hongfu, ZENG Haijun, SUN Jianzhong, YANG Weidong, WEI Fang, CHEN Xuefeng. Current status, challenges and opportunities of civil aero-engine diagnostics & health management Ⅱ: Comprehensive off-board diagnosis, life management and intelligent condition based MRO [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(9): 625574-625574.
[13]	LIU Jiaqi, FENG Yunwen, LU Cheng, XUE Xiaofeng, PAN Weihuang. Safety analysis of aero-engine operation based on intelligent neural network [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(9): 625375-625375.
[14]	SHAO Yiwei, CHEN Jiayu, LIN Cuiying, WAN Cheng, GE Hongjuan, SHI Zhilong. Gearbox fault diagnosis with small training samples: An improved deep forest based method [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(8): 625429-625429.
[15]	KANG Yuxiang, CHEN Guo, WEI Xunkai, ZHOU Lei. Deep residual hedging network and its application in fault diagnosis of rolling bearings [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(8): 625201-625201.

训练周期	准确率/%
100	96.19
150	96.76
200	98.57
250	98.57
300	98.57