Projectile Trajectory Prediction Based on CNN-LSTM Model

doi:10.12382/bgxb.2022.0511

Abstract

Abstract:

To solve the problem of nonlinear trajectory prediction of projectile, a novel hybrid trajectory prediction model based on convolutional neural network (CNN) and long and short-term memory (LSTM) neural network is proposed. A 6DOF projectile movement model is established, and a substantial dataset of trajectory samples is obtained through exterior ballistics simulations employing the four-order Runge-Kutta method. Secondly, the hybrid CNN-LSTM trajectory prediction model is proposed, and the input and output trajectory data pairs are constructed by using the sliding window method and first-order difference method, which transforms the prediction problem into a supervised learning problem. Then, the proposed model is compared with LSTM neural network model, gated recurrent unit (GRU) neural network model and back propagation (BP) neural network model using the same dataset. The results show that the average cumulative prediction error of CNN-LSTM model after 3s is about 14.83m in the x-axis direction, 20.77m in the y-axis direction and 0.75m in the z-axis direction. The trajectory prediction accuracy of CNN-LSTM neural network model is better than that of a single model, which provides valuable insights for advancing projectile trajectory prediction research.

Key words: ballistic model, deep learning, supervised learning, CNN-LSTM neural network model, trajectory prediction

CLC Number:

TJ012.3

ZHENG Zhiwei, GUAN Xueyuan, FU Jian, MA Xunqiong, YIN Shang. Projectile Trajectory Prediction Based on CNN-LSTM Model[J]. Acta Armamentarii, 2023, 44(10): 2975-2983.

Figures/Tables 13

References 23

[1]	张昊, 周军辉, 吴松, 等. 飞航武器遥测系统现状及发展趋势分析[J]. 飞航导弹, 2017(11): 80-84.
	ZHANG H, ZHOU J H, WU S, et al. Analysis of current situation and development trend of aerial weapon telemetry system[J]. Maneuverable Missile, 2017(11): 80-84. (in Chinese)
[2]	邵云峰, 崔玉江. 弹载“黑匣子”设计方法[J]. 现代防御技术, 2016, 44(2): 203-208.
	SHAO Y F, CUI Y J. Design method of black casket in missile[J]. Modern Defence Technology, 2016, 44(2): 203-208. (in Chinese)
[3]	修观, 王良明, 杨荣军. 线性弹道模型建立与仿真[J]. 海军工程大学学报, 2010, 22(2): 84-91, 96.
	XIU G, WANG L M, YANG R J. Construction and simulation of linear trajectory model[J]. Journal of Naval University of Engineering, 2010, 22(2): 84-91, 96. (in Chinese)
[4]	FRESCONI F, COOPER G, COSTELLO M. Practical assessment of real-time impact point estimators for smart weapons[J]. Journal of Aerospace Engineering, 2011, 24(1): 1-11. doi: 10.1061/(ASCE)AS.1943-5525.0000044 URL
[5]	史金光, 刘猛, 曹成壮, 等. 弹道修正弹落点预报方法研究[J]. 弹道学报, 2014, 26(2): 29-33.
	SHI J G, LIU M, CAO C Z, et al. Research on method of falling point prediction for trajectory correction projectile[J]. Journal of Ballistics, 2014, 26(2): 29-33. (in Chinese)
[6]	XING F Y, XIE Y P, SU H, et al. Deep learning in microscopy image analysis: a survey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29(10): 4550-4568. doi: 10.1109/TNNLS.2017.2766168 pmid: 29989994
[7]	XIA Y. Research on statistical machine translation model based on deep neural network[J]. Computing, 2020, 102(2): 643-661. doi: 10.1007/s00607-019-00752-1
[8]	SONG Z J. English speech recognition based on deep learning with multiple features[J]. Computing, 2020, 102(1): 663-682. doi: 10.1007/s00607-019-00753-0
[9]	WANG Z, WANG X Z. A deep stochastic weight assignment network and its application to chess playing[J]. Journal of Parallel and Distributed Computing, 2018, 117: 205-211. doi: 10.1016/j.jpdc.2017.08.013 URL
[10]	HAN P, YUE J C, FANG C, et al. Short-term 4D trajectory prediction based on LSTM neural network[J]. Proceedings of SPIE, 2020, 11427: 2550425.
[11]	QIN W T, TANG J, LU C, et al. Trajectory prediction based on long short-term memory network and Kalman filter using hurricanes as an example[J]. Computational Geosciences, 2021, 25(3): 1005-1023. doi: 10.1007/s10596-021-10037-2
[12]	SHI Z Y, XU M, PAN Q. 4-D flight trajectory prediction with constrained LSTM network[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(11): 7242-7255. doi: 10.1109/TITS.2020.3004807 URL
[13]	张宏鹏, 黄长强, 唐上钦, 等. 基于卷积神经网络的无人作战飞机飞行轨迹实时预测[J]. 兵工学报, 2020, 41(9): 1894-1903. doi: 10.3969/j.issn.1000-1093.2020.09.022
	ZHANG H P, HUANG C Q, TANG S Q, et al. CNN-based real-time prediction method of flight trajectory of unmanned combat aerial vehicle[J]. Acta Armamentarii, 2020, 41(9): 1894-1903. (in Chinese) doi: 10.3969/j.issn.1000-1093.2020.09.022
[14]	WANG R P, CUI Y, SONG X, et al. Multi-information-based convolutional neural network with attention mechanism for pedestrian trajectory prediction[J]. Image and Vision Computing, 2021, 107: 104110. doi: 10.1016/j.imavis.2021.104110 URL
[15]	ZHANG X G, MAHADEVAN S. Bayesian neural networks for flight trajectory prediction and safety assessment[J]. Decision Support Systems, 2020, 131: 113246. doi: 10.1016/j.dss.2020.113246 URL
[16]	XIE G, SHANGGUAN A Q, FEI R, et al. Motion trajectory prediction based on a CNN-LSTM sequential model[J]. Science China Information Sciences, 2020, 63(11): 233-253.
[17]	MA L, TIAN S. A hybrid CNN-LSTM model for aircraft 4D trajectory prediction[J]. IEEE Access, 2020, 8(8): 134668-134680. doi: 10.1109/Access.6287639 URL
[18]	李兴隆, 贾方秀, 王晓鸣, 等. 基于线性弹道模型的末段修正弹落点预测[J]. 兵工学报, 2015, 36(7): 1188-1194. doi: 10.3969/j.issn.1000-1093.2015.07.006
	LI X L, JIA F X, WANG X M, et al. Impact point prediction of terminal correction projectile based on linear trajectory model[J]. Acta Armamentarii, 2015, 36(7): 1188-1194. (in Chinese)
[19]	WERNERT P, THEODOULIS S, MOREL Y. Flight dynamics properties of 155mm spin-stabilized projectiles analyzed in different body frames[C]// Proceedings of Atmospheric Flight Mechanics Conference. Toronto, Ontario, Canada: AIAA, 2010: 1-17.
[20]	王晓鹏, 王雨时, 卢凤生, 等. 155m口径火炮榴弹结构特征数分布特性研究[J]. 探测与控制学报, 2015, 37(5): 66-72.
	WANG X P, WANG Y S, LU F S, et al. Distribution characteristics of 155mm caliber projectile’s structural characteristics parameters[J]. Journal of Detection & Control, 2015, 37(5): 66-72. (in Chinese)
[21]	史继刚, 易文俊, 管军, 等. 基于粒子群-牛顿算法的弹丸阻力系数辨识[J]. 兵器装备工程学报, 2017, 38(2): 23-26, 29.
	SHI J G, YI W J, GUAN J, et al. Drag coefficient identification of spinning projectile using particle swarm Newton iteration method[J]. Journal of Ordnance Equipment Engineering, 2017, 38(2): 23-26, 29. (in Chinese)
[22]	LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324. doi: 10.1109/5.726791 URL
[23]	HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. doi: 10.1162/neco.1997.9.8.1735 pmid: 9377276

参数	采样值	采样间隔
发射角/(°)	[20,66]	2
初始速度/(m·s^-1)	[600,900]	10

参数	采样值	采样间隔
发射角/(°)	[20,66]	2
初始速度/(m·s^-1)	[600,900]	10

层定义	层参数	层输入、输出尺寸
1Dconv层1	卷积核数量10,长度2	输入:(None,10,4)
		输出:(None,9,10)
最大池化层1	池化窗口长度2	输入:(None,9,10)
		输出:(None,4,10)
1Dconv层2	卷积核数量10,长度2	输入:(None,4,10)
		输出:(None,3,10)
最大池化层2	池化窗口长度2	输入:(None,3,10)
		输出:(None,1,10)
LSTM层1	神经元数量30	输入:(None,1,10)
		输出:(None,1,30)
Dropout层1	dropout率0.2	输入:(None,1,30)
		输出:(None,1,30)
LSTM层2	神经元数量30	输入:(None,1,30)
		输出:(None,30)
Dropout层2	dropout率0.2	输入:(None,30)
		输出:(None,30)
全连接层	神经元数量3	输入:(None,30)
		输出:(None,3)

层定义	层参数	层输入、输出尺寸
1Dconv层1	卷积核数量10,长度2	输入:(None,10,4)
		输出:(None,9,10)
最大池化层1	池化窗口长度2	输入:(None,9,10)
		输出:(None,4,10)
1Dconv层2	卷积核数量10,长度2	输入:(None,4,10)
		输出:(None,3,10)
最大池化层2	池化窗口长度2	输入:(None,3,10)
		输出:(None,1,10)
LSTM层1	神经元数量30	输入:(None,1,10)
		输出:(None,1,30)
Dropout层1	dropout率0.2	输入:(None,1,30)
		输出:(None,1,30)
LSTM层2	神经元数量30	输入:(None,1,30)
		输出:(None,30)
Dropout层2	dropout率0.2	输入:(None,30)
		输出:(None,30)
全连接层	神经元数量3	输入:(None,30)
		输出:(None,3)

模型	层定义	层参数
	LSMT层1	神经元数量50
	Dropout层1	Dropout率0.2
LSTM模型	LSMT层2	神经元数量50
	Dropout层2	Dropout率0.2
	全连接层	神经元数量3
	GRU层1	神经元数量50
	Dropout层1	Dropout率0.2
GRU模型	GRU层2	神经元数量50
	Dropout层2	Dropout率0.2
	全连接层	神经元数量3
	全连接层1	神经元数量100
	Dropout层1	Dropout率0.2
BP模型	全连接层2	神经元数量100
	Dropout层2	Dropout率0.2
	全连接层	神经元数量3