基于Mogrifier-BiGRU的飞行器轨迹预测

doi:10.12382/bgxb.2022.0750

摘要/Abstract

摘要：

针对当前飞行器轨迹预测准确性低的问题,引入双向传播机制和Mogrifier数据耦合模块,改进传统门控循环单元网络,提出基于Mogrifier-BiGRU的飞行器轨迹预测算法,加强网络对历史数据的学习与记忆,使得输入信息与隐藏层数据的充分耦合,提高预测准确度。仿真结果表明,所提方法对飞行器轨迹预测的准确度可达到96.26%,满足我方作战指挥人员对战场态势趋势准确预测的实际需求。

关键词: 轨迹预测, 双向门控循环单元, Mogrifier数据模块, 深度学习

Abstract:

For the low accuracy of trajectory prediction of current aircraft, the bidirectional propagation mechanism and Mogrifier data coupling module are introduced to improve the gated recurrent unit (GRU) network. A prediction algorithm of aircraft trajectory based on Mogrifier-BiGRU is proposed, which strengthens the learning and memory of historical data, makes the input information fully coupled with the hidden layer data, and improves the prediction accuracy. The simulated results show that the accuracy of the proposed method for predicting the aircraft maneuver trajectory can reach 96.26%, which meets the actual demand of combat commanders for accurate prediction of battlefield situation.

Key words: trajectory prediction, bidirectional gated recurrent unit, Mogrifier data module, deep learning

中图分类号:

E926

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.

ZHANG Kun, DU Ruiyi, SHI Haotian, HUA Shuai. Prediction of Aircraft Trajectory Based on Mogrifier-BiGRU[J]. Acta Armamentarii, 2024, 45(2): 373-384.

图/表 19

图1 飞行器惯性坐标系与机体坐标系模型

Fig.1 Inertial coordinate frame model and body frame model of aircraft

图2 Mogrifier-BiGRU网络模型

Fig.2 Mogrifier-BiGRU network model

图3 BiGRU网络示意图

Fig.3 Schematic diagram of BiGRU network

图4 Mogrifier数据耦合单元结构

Fig.4 Mogrifier data coupling unit structure

图5 网络训练流程图

Fig.5 Network training flow chart

图6 不同耦合参数网络模型准确率对比

Fig.6 Comparison of accuracy ratios of network model with different coupling parameters

图7 不同耦合参数网络模型偏差率对比

Fig.7 Comparison of deviation ratios of network model with different coupling parameters

表1 网络属性对比表

Table 1 Comparison table of network attributes

模型	r	平均准确率/%	平均偏差率/%
	0	86.24	13.81
	1	88.50	10.97
	2	86.18	14.96
	3	88.98	12.35
	4	86.61	14.92
Mogrifier_BiGRU	5	87.70	13.16
	6	88.25	11.10
	7	87.95	13.04
	8	88.50	12.33
	9	86.95	12.25
	10	86.48	14.97
	20	83.64	13.98
Multiplicative_BiGRU		86.13	13.13

表2 网络参数汇总表

Table 2 Summary table of network parameters

网络参数	取值	网络参数	取值
飞行器轨迹数据量	1000	训练回合	50000
η	0.8	ε	0.05
采样时间间隔/s	0.25	GRU网络层数	3
r	1	GRU网络隐藏节点数	256

图8 网络准确率、偏差率、损失函数对比

Fig.8 Network accuracy, deviation rate, loss function comparison diagram

表3 飞行器轨迹预测准确率、偏差率对比

Table 3 Comparison of accuracy and deviation rate of aircraft trajectory prediction

网络	平均准确率/%	平均偏差率/%
GRU	26.50	20.19
BiGRU	85.93	12.05
BiLSTM	85.20	12.17
Multiplicative_BiGRU	91.42	8.20
Mogrifier_BiGRU	94.40	4.74

表4 飞行器轨迹预测损失函数对比

Table 4 Comparison of experimental loss functions for prediction of aircraft trajectory

网络	RMSE	MAE
GRU	0.0160	0.0090
BiGRU	0.0053	0.0033
BiLSTM	0.0051	0.0031
Multiplicative_BiGRU	0.0046	0.0025
Mogrifier_BiGRU	0.0037	0.0022

图9 网络准确率、偏差率、损失函数对比

Fig.9 Comparison diagram of network accuracies, deviation rates and loss functions

表5 飞行器轨迹与姿态预测实验A结果对比

Table 5 Comparison of the results of trajectory and attitude prediction experiments A

网络	平均准确率/%	平均偏差率/%
GRU	15.17	21.85
BiGRU	83.88	8.20
BiLSTM	84.98	8.29
Multiplicative_BiGRU	88.88	6.05
Mogrifier_BiGRU	90.32	5.78

图10 网准确率、偏差率、损失函数对比

Fig.10 Comparison of network accuracies, deviation rates and loss functions

表6 飞行器轨迹与姿态预测实验B结果对比

Table 6 Comparison of the results of trajectory and attitude prediction experiments B

网络	平均准确率/%	平均偏差率/%
GRU	84.29	8.21
BiGRU	90.93	6.60
BiLSTM	91.88	5.42
Multiplicative_BiGRU	94.71	2.99
Mogrifier_BiGRU	96.26	3.85

表7 网络训练及预测时间表

Table 7 Network training and prediction schedule

仿真实验	训练时长/s	预测时间/s
Mogrifier-BiGRU网络预测实验	1243.67	0.0097
飞行器轨迹与姿态预测实验A	1288.23	0.0101
飞行器轨迹与姿态预测实验B	2963.59	0.0242

表8 飞行器轨迹与姿态预测实验损失函数对比

Table 8 Comparison of experimental loss functions for prediction of aircraft trajectory and attitude

实验	网络	RMSE	MAE
	GRU	0.0208	0.0132
	BiGRU	0.0067	0.0045
实验A	BiLSTM	0.0068	0.0041
	Multiplicative_BiGRU	0.0063	0.0043
	Mogrifier_BiGRU	0.0061	0.0040
	GRU	0.0131	0.0073
	BiGRU	0.0035	0.0023
实验B	BiLSTM	0.0032	0.0021
	Multiplicative_BiGRU	0.0026	0.0015
	Mogrifier_BiGRU	0.0019	0.0012

图11 实验结果

Fig.11 Experimental results

参考文献 27

[1]	奚之飞, 徐安, 寇英信, 等. 基于前景理论的空战目标威胁评估[J]. 兵工学报, 2020, 41(6):1236-1248. doi: 10.3969/j.issn.1000-1093.2020.06.021
	XI Z F, XU A, KOU Y X, et al. Air combat target threat assessment based on prospect theory[J]. Acta Armamentarii, 2020, 41(6): 1236-1248. (in Chinese) doi: 10.3969/j.issn.1000-1093.2020.06.021
[2]	赵梓烨, 刘海鸥, 陈慧岩. 分布式电驱动无人高速履带车辆越野环境轨迹预测方法研究[J]. 兵工学报, 2019, 40(4):680-688. doi: 10.3969/j.issn.1000-1093.2019.04.002
	ZHAO Z Y, LIU H O, CHEN H Y. Research on trajectory prediction method of distributed high speed electric drive unmanned tracked vehicle in off-road conditions[J]. Acta Armamentarii, 2019, 40(4):680-688. (in Chinese) doi: 10.3969/j.issn.1000-1093.2019.04.002
[3]	LAI H Y, KE H Y. Projectile flight trajectory and position estimation system based on stereo vision[J]. Sensors and Materials, 2019, 31(11): 3483-3493. doi: 10.18494/SAM.2019.2485 URL
[4]	乔少杰, 韩楠, 朱新文, 等. 基于卡尔曼滤波的动态轨迹预测算法[J]. 电子学报, 2018, 46(2):418-423. doi: 10.3969/j.issn.0372-2112.2018.02.022
	QIAO S J, HAN N, ZHU X W, et al. A dynamic trajectory prediction algorithm based on kalman filter[J]. Acta Electronica Sinica, 2018, 46(2):418-423. (in Chinese)
[5]	PANG Y T, ZHAO X Y, YAN H, et al. Data-driven trajectory prediction with weather uncertainties: a bayesian deep learning approach[J]. Transportation Research Part C: Emerging Technologies, 2021, 130(1):103326. doi: 10.1016/j.trc.2021.103326 URL
[6]	张凯, 熊家军, 李凡, 等. 基于意图推断的高超声速滑翔目标贝叶斯轨迹预测[J]. 宇航学报, 2018, 39(11):1258-1265.
	ZHANG K, XIONG J J, LI F, et al. Bayesian trajectory prediction for a hypersonic gliding reentry vehicle based on intent inference[J]. Journal of Astronautics, 2018, 39(11):1258-1265. (in Chinese)
[7]	SONG S N, ZHANG Y J. Driving trajectory prediction method based on adaboost-markov model optimization[C]//Proceedings of the 2019 2nd International Conference on Mechanical Engineering,Industrial Materials and Industrial Electronics. Dalian, China: Clausius Scientific Press, 2019: 362-368.
[8]	DING J H, LIU H Z, YANG L T, et al. Multi-user multivariate multi-order markov based multi-modal user mobility pattern prediction[J]. IEEE Internet of Things Journal, 2020, 7(5): 4519-4531. doi: 10.1109/JIoT.6488907 URL
[9]	MA Z M, ZHONG H W, XIE L, et.al. Month ahead average daily electricity price profile forecasting based on a hybrid nonlinear regression and SVM model: an ERCOT case study[J]. Journal of Modern Power Systems and Clean Energy, 2018, 6(2): 281-291. doi: 10.1007/s40565-018-0395-3
[10]	PRADA J, DORRONSORO J R. General noise support vector regression with non-constant uncertainty intervals for solar radiation prediction[J]. Journal of Modern Power Systems and Clean Energy, 2018, 6(2): 268-280. doi: 10.1007/s40565-018-0397-1
[11]	CASADO E, CIVITA M L, VILAPLANA M, et al. Quantification of aircraft trajectory prediction uncertainty using polynomial chaos expansions[C]//Proceedings of the 2017 IEEE/AIAA 36th Digital Avionics Systems Conference. St. Petersburg, FL,US:IEEE, 2017: 1-11.
[12]	KARLAFTIS M G, VLAHOGIANNI E I. Statistical methods versus neural networks in transportation research: differences, similarities and some insights[J]. Transportation Research Part C: Emerging Technologies, 2011, 19(3): 387-399. doi: 10.1016/j.trc.2010.10.004 URL
[13]	WANG M, WU J W, GUO J H, et al. Multi-point prediction of aircraft motion trajectory based on ga-elman-regularization neural network[J]. Integrated Ferroelectrics, 2020, 210(1): 115-126.
[14]	张涛, 郭基联, 徐西蒙, 等. 基于Elman神经网络的战斗机空战轨迹预测[J]. 飞行力学, 2018, 36(5):86-91.
	ZHANG T, GUO J L, XU X M, et al. Air combat trajectory prediction of a fighter based on Elman neural network[J]. Flight Dynamics, 2018, 36(5):86-91. (in Chinese)
[15]	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 S 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
[16]	JIAO J C, CHEN X P, GUAN M, et al. TCL: a taxi trajectory prediction model combining time and space features[J]. The Journal of China Universities of Posts and Telecommunications, 2021, 28(3):63-75.
[17]	LI T, NI A N, ZHANG C Q, et al. Short-term traffic congestion prediction with Conv-BiLSTM considering spatio-temporal features[J]. IET Intelligent Transport Systems, 2021, 14(14): 1978-1986. doi: 10.1049/itr2.v14.14 URL
[18]	李松, 汪圣利. 基于BP神经网络的非线性滤波算法研究[J]. 电子测量技术, 2018, 41(12):34-39.
	LI S, WANG S L. Research of nonlinear filtering algorithm based on BP neural network[J]. Electronic Measurement Technology, 2018, 41(12):34-39. (in Chinese)
[19]	PARK J, JEONG J, PARK Y. Ship trajectory prediction based on Bi-LSTM using spectral-clustered ais data[J]. Journal of Marine Science and Engineering, 2021, 9(9):1037. doi: 10.3390/jmse9091037 URL
[20]	MA L, TIAN S. A hybrid CNN-LSTM model for aircraft 4D trajectory prediction[J]. IEEE Access, 2020, 8: 134668-134680. doi: 10.1109/Access.6287639 URL
[21]	张永梅, 赖裕平, 马健喆, 等. 基于视频的装甲车和飞机检测跟踪及轨迹预测算法[J]. 兵工学报, 2021, 42(3):545-554. doi: 10.3969/j.issn.1000-1093.2021.03.010
	ZHANG Y M, LAI Y P, MA J Z, et al. A video-based prediction algorithm for armored vehicle and aircraft detection/tracking and trajectory[J]. Acta Armamentarii, 2021, 42(3):545-554. (in Chinese) doi: 10.3969/j.issn.1000-1093.2021.03.010
[22]	CHUNG J, GILCEHRE C, CHO K H, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling: arXiv:1412.3555[R/OL]. Ithaca, NY, US: Cornell University, 2014:1412.3555. https://arxiv.org/abs/1412.3555.
[23]	张宏鹏, 黄长强, 轩永波, 等. 用门控循环单元实时预测空战飞行轨迹[J]. 系统工程与电子技术, 2020, 42(11):2546-2552. doi: 10.3969/j.issn.1001-506X.2020.11.17
	ZHANG H P, HUANG C Q, XUAN Y B, et al. Real-time prediction of air combat flight trajectory using GRU[J]. Systems Engineering and Electronics, 2020, 42(11):2546-2552. (in Chinese) doi: 10.3969/j.issn.1001-506X.2020.11.17
[24]	张宗腾, 张琳, 汪文峰, 等. 一种基于双向GRU的UAV飞行轨迹预测方法[J]. 电光与控制, 2022, 29(3):11-15, 26.
	ZHANG Z T, ZHANG L, WANG W F, et al. A method for UAV flight trajectory prediction based on bidirectional GRU[J]. Electronics Optics & Control, 2022, 29(3):11-15, 26. (in Chinese)
[25]	MELIS G, KOČISKÝ T, BLUNSOM P. Mogrifier LSTM: arXiv:1909.01792[R/OL]. Ithaca, NY, US: Cornell University, 2019:1909.01792. https://arxiv.org/abs/1909.01792.
[26]	KINGMA D, BA J. Adam: a method for stochastic optimization: arXiv:1412.6980[R/OL]. Ithaca, NY, US: Cornell University, 2014:1412.6980. https://arxiv.org/abs/1412.6980.
[27]	KRAUSE B, LU L, MURRAY I, et al. Multiplicative LSTM for sequence modelling: arXiv:1609.07959[R/OL]. Ithaca, NY, US: Cornell University, 2016:1609.07959. https://arxiv.org/abs/1609.07959.