基于神经网络的高超声速飞行器惯导系统精度提高方法

doi:10.12305/j.issn.1001-506X.2022.04.28

摘要/Abstract

摘要：

针对目前惯性系统误差补偿模型对静态误差和动态误差处理能力不足的问题, 为适应高超声速飞行器长航时、高精度的惯性导航要求, 基于神经网络提出一种加速度计拟合模型。在高超声速飞行器飞行前期有准确的卫星导航信息时, 收集导航信息和加速度计脉冲信息, 利用神经网络强大的非线性拟合能力, 在飞行过程中进行在线训练, 得到精确的惯性系统模型。仿真结果表明, 在存在逐次通电误差和不考虑二次项误差系数的误差补偿模型方法位置导航偏差在数公里和数百米量级的情况下, 相同时间内所提方法的位置导航偏差仅为数十米量级, 有效提高了高超声速飞行器的导航精度。

关键词: 神经网络, 捷联惯性系统, 加速度计, 高超声速飞行器, 误差补偿模型

Abstract:

Aiming at the problem that the current inertial system error compensation model is insufficient in processing static and dynamic errors, in order to meet the requirements of long-endurance and high-precision inertial navigation of hypersonic aircraft, an accelerometer fitting model based on neural network is proposed. When the hypersonic vehicle has accurate satellite navigation information in the early stage of flight, the navigation information and accelerometer pulse information are collected, and the powerful nonlinear fitting ability of the neural network is used to conduct online training during the flight to obtain an accurate inertial system model. The simulation results show that when the power-on error and the error compensation model method does not consider the quadratic term error coefficient, the position navigation deviation is on the order of several kilometers and hundreds of meters, the position navigation deviation of the proposed method is only on the order of tens of meters in the same time, which effectively improve the navigation accuracy of the hypersonic vehicle.

Key words: neural network, strapdown inertial navigation system (SINS), accelerometer, hypersonic vehicle, error compensation model

中图分类号:

V249.32

胥涯杰, 鲜勇, 李邦杰, 任乐亮, 李少朋, 郭玮林. 基于神经网络的高超声速飞行器惯导系统精度提高方法[J]. 系统工程与电子技术, 2022, 44(4): 1301-1309.

Yajie XU, Yong XIAN, Bangjie LI, Leliang REN, Shaopeng LI, Weilin GUO. Method for improving the precision of hypersonic vehicle inertial navigation system based on neural network[J]. Systems Engineering and Electronics, 2022, 44(4): 1301-1309.

图/表 15

图1

图2

表1

图3

表2

图4

图5

图6

表3

表4

表5

表6

图7

表7

表8

参考文献 33

1	SDG A , DQ B , GP C , et al. An aerothermodynamic design optimization framework for hypersonic vehicles[J]. Aerospace Science and Technology, 2019, 84, 339- 347. doi: 10.1016/j.ast.2018.09.042
2	LUO C X , LEI H M , LI J , et al. A new adaptive neural control scheme for hypersonic vehicle with actuators multiple constraints[J]. Nonlinear Dynamics, 2020, 100 (3): 3529- 3553.
3	LI K L , HU B Q , CHANG L B , et al. Comparison of direct navigation mode and indirect navigation mode for integrated SINS/GPS[J]. Transactions of the Institute of Measurement and Control, 2016, 38 (1): 3- 13. doi: 10.1177/0142331214568236
4	LIU P J , WANG B , HOU D D , et al. An approach for DVL-aided SINS in-motion alignment based on observability analysis[J]. IEEE Sensors Journal, 2021, 21 (15): 17131- 17143. doi: 10.1109/JSEN.2021.3081865
5	AKSONENKO P, AVRUTOV V, LAZAREV Y F, et al. Overclocking algorithms for SINS[C]//Proc. of the IEEE International Conference Actual Problems of Unmanned Aerial Vehicles Developments, 2018.
6	TAO Y , RUI X , YANG F F , et al. Development of a MRE isolation system for strapdown inertial measurement unit[J]. Mechanical Systems and Signal Processing, 2019, 117 (2): 553- 568.
7	KARINCK D, BA L G, KOLAND L, et al. Honeywell gun-hard inertial measurement unit (IMU) development[C]//Proc. of the Position Location and Navigation Symposium, 2004.
8	石文峰, 王省书, 郑佳兴, 等. 激光陀螺捷联惯导系统多位置系统级标定方法[J]. 红外与激光工程, 2016, 45 (11): 99- 106.
	SHI W F , WANG S S , ZHENG J X , et al. Multi-position systematic calibration method for RLG-SINS[J]. Infrared and Laser Engineering, 2016, 45 (11): 99- 106.
9	王子卉, 程向红, 范时秒. 基于双轴转台的捷联惯性导航系统8位置系统级标定方法[J]. 中国惯性技术学报, 2019, 27 (1): 23- 31.
	WANG Z H , CHENG X H , FAN S M . Eight-position systema-tic calibration method for SINS based on two-axis turntable with temperature compensation[J]. Journal of Chinese Inertial Technology, 2019, 27 (1): 23- 31.
10	DAI M M , ZHANG C X , LU J Z . In-field calibration method for DTG IMU including g-sensitivity biases[J]. IEEE Sensors Journal, 2019, 19 (13): 4972- 4981. doi: 10.1109/JSEN.2019.2898104
11	DONG C M , REN S Q , CHEN X J , et al. A separated calibration method for inertial measurement units mounted on three-axis turntables[J]. Sensors, 2018, 18 (9): 2846. doi: 10.3390/s18092846
12	杨广强, 王志涛, 梁化勇, 等. 基于转台误差隔离的捷联惯导系统尺寸效应参数标定[J]. 中国惯性技术学报, 2020, 28 (1): 15- 19.
	YANG G Q , WANG Z T , LIANG H Y , et al. Calibration for size effect parameters of SINS based on turntable error isolation[J]. Journal of Chinese Inertial Technology, 2020, 28 (1): 15- 19.
13	KLIMKOVICH B V , TOLOCHKO A M . Consideration for size effect in SINS calibration[J]. Gyroscopy and Navigation, 2015, 6 (3): 230- 235. doi: 10.1134/S2075108715030074
14	GAO P Y , LI K , SONG T X , et al. An accelerometers size effect self-calibration method for tri-axis rotational inertial navigation System[J]. IEEE Trans.on Industrial Electronics, 2017, 65 (2): 1644- 1655.
15	孙闯, 任顺清, 王振桓. 加速度计在精密离心机上的标定方法与误差分析[J]. 中国惯性技术学报, 2019, 27 (1): 121- 128.
	SUN C , REN S Q , WANG Z H . Calibration method of accelero-meter on precision centrifuge and its error analysis[J]. Journal of Chinese Inertial Technology, 2019, 27 (1): 121- 128.
16	NING X L , YUAN W P , LIU Y H . A tightly coupled rotational SINS/CNS integrated navigation method for aircraft[J]. Journal of Systems Engineering and Electronics, 2019, 30 (4): 770- 782. doi: 10.21629/JSEE.2019.04.14
17	WANG D J , LV H F , AN X Y , et al. A high-accuracy constrained SINS/CNS tight integrated navigation for high-orbit automated transfer vehicles[J]. Acta Astronaut, 2018, 151, 614- 625. doi: 10.1016/j.actaastro.2018.07.015
18	OUYANG W, WU Y X, CHEN H Y. INS/Odometer land navigation by accurate measurement modeling and multiple-model adaptive estimation[C]//Proc. of the 27th Saint Petersburg International Conference on Integrated Navigation Systems, 2020.
19	GORKEM S , BILLUR B . Improvements in deterministic error modeling and calibration of inertial sensors and magnetometers[J]. Sensors and Actuators, A. Physical, 2016, 247, 522- 538. doi: 10.1016/j.sna.2016.06.024
20	ZHAI H Q , WANG L H . The robust residual-based adaptive estimation Kalman filter method for strap-down inertial and geomagnetic tightly integrated navigation system[J]. Review of Scientific Instruments, 2020, 91 (10): 104501. doi: 10.1063/5.0019305
21	XIA X W , SUN Q . An on-line calibration method of star sensor/inertial navigation system for marine[J]. IAENG Internaitonal Journal of Computer Science, 2020, 47 (1): 19- 24.
22	DU H S, QIAN S, ZHANG Y, et al. Inertial navigation system positioning assisted by star sensor[C]//Proc. of the IEEE International Conference on Control, 2017.
23	SHI L , YANG H T , BAI B W , et al. Transmission channel characteristics of relay dual-polarization MIMO system for hypersonic vehicles under plasma sheath[J]. IEEE Trans.on Plasma Science, 2018, 46 (4): 917- 927. doi: 10.1109/TPS.2018.2811547
24	BELHAJEM I, MAISSA Y B, TAMTAOUI A. A hybrid low cost approach using extended Kalman filter and neural networks for real time positioning[C]//Proc. of the IEEE International Conference on Information Technology for Organizations Deve-lopment, 2016.
25	郭玮林, 鲜勇, 张大巧, 等. 高超声速飞行器惯导系统误差参数两次优化辨识方法[J]. 控制与决策, 2020, 35 (1): 25- 34.
	GUO W L , XIAN Y , ZHANG D Q , et al. Quadratic optimization identification research on error parameters of SINS for hypersonic vehicle[J]. Control and Decision, 2020, 35 (1): 25- 34.
26	ZHANG T , ZHU Y Y , ZHOU F , et al. Coarse alignment technology on moving base for SINS based on the improved quaternion filter algorithm[J]. Sensors, 2017, 17 (6): 1424. doi: 10.3390/s17061424
27	QIU Z B , QIAN H M . Modified multiplicative quaternion cubature Kalman filter for attitude estimation[J]. International Journal of Adaptive Control and Signal Processing, 2018, 32 (8): 1182- 1190. doi: 10.1002/acs.2895
28	REGIMANU B , DAS I C , RAO K S . Development of multistage digital filters for dither signal removal in ring laser gyro[J]. Frequenz, 2019, 73 (3-4): 123- 130. doi: 10.1515/freq-2018-0195
29	SCHWARTZ S, GUTTY F, FE U G, et al. Solid-state ring laser gyro for aerospace applications[C]//Proc. of the Society of Photo-optical Instrumentation Engineers Conference Series, 2017.
30	鲜勇, 李邦杰, 雷刚, 等. 弹道导弹精度分析方法[M]. 长沙: 国防科技大学出版社, 2012: 93- 94.
	XIAN Y , LI B J , LEI G , et al. Ballistic missile precision analy-sis method[M]. Changsha: National University of Defense Technology Press, 2012: 93- 94.
31	ZOUHAIR C , NOUREDDINE A , KHALID M , et al. A novel architecture combined with optimal parameters for back propagation neural networks applied to anomaly network intrusion detection[J]. Computers & Security, 2018, 75 (6): 36- 58.
32	RUBIO J . Stability analysis of the modified Levenberg-Marquardt algorithm for the artificial neural network training[J]. IEEE Trans.on Neural Networks and Learning Systems, 2021, 32 (8): 3510- 3524. doi: 10.1109/TNNLS.2020.3015200
33	赵林东. 高超声速飞行器建模及巡航跟踪控制技术研究[D]. 北京: 北京理工大学, 2015.
	ZHAO L D. Research on modeling and tracking control technology for hypersonic vehicle during cruise phase[D]. Beijing: Beijing Institute of Technology, 2015.

节点数	训练误差
节点数	X方向	Y方向	Z方向
4	5.50×10^-6	8.33×10^-7	3.45×10^-7
5	5.08×10^-6	1.52×10^-6	2.02×10^-7
6	1.54×10^-6	4.04×10^-7	9.37×10^-8
7	2.30×10^-4	5.44×10^-5	7.78×10^-6
8	2.85×10^-5	4.58×10^-5	1.95×10^-5
9	3.53×10^-6	8.17×10^-7	6.26×10^-8
10	1.47×10^-3	2.23×10^-5	2.28×10^-5
11	5.24×10^-5	1.47×10^-5	2.61×10^-6
12	1.97×10^-4	2.26×10^-5	2.15×10^-7
13	8.59×10^-4	4.42×10^-5	2.86×10^-6
14	8.61×10^-4	1.62×10^-5	5.82×10^-6

误差参数	漂移值	单位
ΔK_0x	5×10^-4	g
ΔK_0y	5×10^-4	g
ΔK_0z	5×10^-4	g
ΔK_1x	1×10^-3	-
ΔK_1y	1×10^-4	-
ΔK_1z	1×10^-4	-

偏差/m	δX	偏差/m	δY	偏差/m	δZ
< 100	9	< 100	24	< 100	13
< 500	48	< 500	90	< 500	62
< 1 000	73	< 700	98	< 1 000	95
< 2 000	100	< 800	100	< 1 500	100

统计项	δX	δY	δZ
平均值	6.45×10²	2.59×10²	4.51×10²
最大值	1.98×10³	7.81×10²	1.46×10³
标准差	4.65×10²	1.85×10²	3.12×10²

偏差/m	δX	偏差/m	δY	偏差/m	δZ
< 10	5	< 10	12	< 10	33
< 100	79	< 100	82	< 50	93
< 200	98	< 200	99	< 100	99
< 400	100	< 300	100	< 200	100