基于贝叶斯神经网络的船用惯导定位修正方法

doi:10.12305/j.issn.1001-506X.2024.04.28

摘要/Abstract

摘要：

船用惯性导航系统(inertial navigation system, INS)通常采用与全球卫星导航系统(global navigation satellite system, GNSS)组合导航的方式提高其长时间工作的定位精度。当GNSS失效时, 其定位误差将随时间迅速发散。针对这一问题, 设计了采用反向传播神经网络(back propagate neural network, BPNN)、根据INS原始输出数据拟合修正经纬度的定位修正方案, 提出了基于Bayesian算法更新网络权重系数的方法, 结合理论分析和试验研究确定了神经元个数与训练数据集的分配方案。实船试验结果表明, 当GNSS失效时, 在后续2 h, 通过24 h历史数据训练得到的神经网络修正INS位置, 相比INS独立工作时的定位误差, 修正后误差均值下降了63%, 误差最大值下降约50%, 最小值下降至0。

关键词: 惯性导航系统, 全球卫星导航系统失效, 反向传播神经网络, Bayesian算法, 定位误差

Abstract:

Ship inertial navigation system (INS) usually integrate with global navigation satellite system (GNSS) to improve its long term position performance. Once GNSS disables, the position error will diverge fast with time. In order to improve INS' long term position accuracy while GNSS disables, a position correction scheme is proposed using back propagation neural network (BPNN) to fit and correct longitude and latitude based on the INS' origin data, and the network's weigh value is updated based on the Bayesian algorithm. While according to the general theory to calculate the number of neurons, the best number of neurons and training sample distribution are determined via a number of experiments. The real ship test results show that while GNSS disables, in the next two hours, neural net correction inertial navigation position obtained by historical data of 24 h is more accurate than the independent-working INS' correction inertial navigation position, the position mean error decrease 63 percent, the maximum error decrease 50 percent and the minimal error decrease to 0.

Key words: inertial navigation system (INS), global navigation satellite system (GNSS) disable, back propagation neural network (BPNN), Bayesian algorithm, position error

中图分类号:

TN911

周红进, 宋辉, 范文良, 王苏, 谷东亮. 基于贝叶斯神经网络的船用惯导定位修正方法[J]. 系统工程与电子技术, 2024, 46(4): 1393-1400.

Hongjin ZHOU, Hui SONG, Wenliang FAN, Su WANG, Dongliang GU. Ship inertial navigation system position correction method based on Bayesian neural network[J]. Systems Engineering and Electronics, 2024, 46(4): 1393-1400.

图/表 12

图1

图2

图3

图4

图5

图6

表1

图7

表2

图8

图9

表3

参考文献 26

1	王思远, 刘海桥, 罗世彬, 等.基于高精度MEMS-INS的双轴旋转调制技术[J].传感器与微系统,2021,40(2):8-11.
	WANG S Y , LIU H Q , LUO S B , et al.Dual-axis rotation modulation technology based on high precision MEMS-INS[J].Transducer and Microsystem Technologies,2021,40(2):8-11.
2	熊崴, 蔡智渊, 齐宇轩, 等.H调制陀螺监控高精度惯性导航系统[J].中国惯性技术学报,2010,18(4):387-395.
	XIONG W , CAI Z Y , QI Y X , et al.High-accuracy inertial navigation system with H-modulation gyro monitoring[J].Journal of Chinese Inertial Technology,2010,18(4):387-395.
3	崔加瑞, 吴文启, 马铁锋, 等.双惯导联合旋转调制光纤陀螺标度因数误差自校正方法[J].中国惯性技术学报,2022,30(5):561-568.
	CUI J R , WU W Q , MA T F , et al.Self-calibration of scale factor error for fiber optic gyroscope based on joint rotation modulation method of dual INS[J].Journal of Chinese Inertial Techno-logy,2022,30(5):561-568.
4	胡佩达, 高钟毓, 吴秋平, 等.壳体翻滚失准角模型最小二乘辨识方法[J].中国惯性技术学报,2009,17(4):393-396.
	HU P D , GAO Z Y , WU Q P , et al.Least square identification method of the misalignment angle of gyro case rotation axis[J].Journal of Chinese Inertial Technology,2009,17(4):393-396.
5	黄凤荣, 朱雨晨, 杨泽清, 等.基于高斯混合模型的惯导/计程仪组合导航方法[J].中国惯性技术学报,2019,27(1):32-35.
	HUANG F R , ZHU Y C , YANG Z Q , et al.SINS/EML navigation method based on Gaussian mixtures unscented Kalman filter[J].Journal of Chinese Inertial Technology,2019,27(1):32-35.
6	LI J , GU M Y , ZHU T L , et al.Research on error correction technology in underwater SINS/DVL integrated positioning and navigation[J].Sensors,2023,23(10):4700-4717. doi: 10.3390/s23104700
7	LIU J X , KE D , PENG H , et al.A tight coupling algorithm for strapdown inertial navigation system/global positioning system adaptive integrated navigation based on variational Bayesian[J].Sustainability,2023,15(16):12477-12489. doi: 10.3390/su151612477
8	MU P C , ZHANG X , QIN P , et al.A variational Bayesian-based simultaneous localization and mapping method for autonomous underwater vehicle navigation[J].Journal of Marine Science and Engineering,2022,10(10):1563-1577. doi: 10.3390/jmse10101563
9	LAI X , TONG S Q , ZHU G R I .Adaptive fuzzy neural network-aided progressive Gaussian approximate filter for GPS/INS integration navigation[J].Measurement,2022,20(10):1641-1656.
10	付强文, 秦永元, 李四海.速度约束辅助车载捷联惯导系统零速校正算法[J].系统工程与电子技术,2013,35(8):1723-1728.
	FU Q W , QIN Y Y , LI S H .ZUPT method for vehicular SINS aided by velocity constrain[J].System Engineering and Electronics,2013,35(8):1723-1728.
11	SEMENIUK L , NOURELDIN A .Bridging GPS outages using neural network estimates of INS position and velocity errors[J].Measurement Science & Technology,2006,17(10):2783.
12	JWO D , BISWAL A , MIR I A , et al.Artificial neural networks for navigation systems: a review of recent research[J].Applied Sciences,2023,13(7):4475-4486. doi: 10.3390/app13074475
13	CHEN X , SHEN C , ZHANG W B , et al.Novel hybrid of strong tracking Kalman filter and wavelet neural network for GPS/INS during GPS outages[J].Measurement,2013,46(10):3847-3854. doi: 10.1016/j.measurement.2013.07.016
14	XU Y , WANG K , JIANG C H , et al.Motion-constrained GNSS/INS integrated navigation method based on BP neural network[J].Remote Sensing,2023,15(1):154-167.
15	WANG G C , XU X S , TONG J W .A novel BPNN-based method to overcome the GPS outages for INS/GPS system[J].IEEE Access,2019,7,82134-82143. doi: 10.1109/ACCESS.2019.2922212
16	SPECHT M .Consistency analysis of global positioning system position errors with typical statistical distributions[J].Journal of Navigation,2021,74(6):1201-1218. doi: 10.1017/S0373463321000485
17	SAADEDDIN K , ABDEL-HAFEZ M F , JARADAT M A , et al.Optimization of intelligent approach for low-cost INS/GPS navigation system[J].Journal of Intelligent & Robotic Systems,2014,73(1/4):325-348.
18	WANG L H , YE W J , ZHOU Y T .Optimal parameters selection of back propagation algorithm in the feed forward neural network[J].Engineering Analysis with Boundary Elements,2023,151(13):575-596.
19	SILVA M E , PEREIRA I , MCCABE B .Bayesian outlier detection in non-Gaussian autoregressive time series[J].Journal of Time Series Analysis,2019,40(5):631-648. doi: 10.1111/jtsa.12439
20	陈怀宇, 尹达一, 张泉, 等.LSTM网络提高MEMS惯导定位精度的分析及验证[J].中国惯性技术学报,2018,26(5):610-615.
	CHEN H Y , YIN D Y , ZHANG Q , et al.Analysis and verification on improving MEMS navigation accuracy based on LSTM network[J].Journal of Chinese Inertial Technology,2018,26(5):610-615.
21	李宇寰, 杨功流, 于沛, 等.基于Bagging模型的惯导系统误差抑制方法[J].中国惯性技术学报,2017,25(1):63-66.
	LI Y H , YANG G L , YU P , et al.Error restraining method for SINS based on Bagging model[J].Journal of Chinese Inertial Technology,2017,25(1):63-66.
22	黄凤荣, 羿博珩, 王旭, 等.基于深度学习与运动状态识别的车辆惯性导航方法[J].中国惯性技术学报,2022,30(5):569-575.
	HUANG F G , YI B H , WANG X , et al.Vehicle inertial navigation method based on deep learning and motion constraints[J].Journal of Chinese Inertial Technology,2022,30(5):569-575.
23	胥涯杰, 鲜勇, 李邦杰, 等.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.
	XU Y J , XIAN Y , LI B J , et al.Method for improving the precision of hypersonic vehicle inertial navigation system based on neural network[J].System Engineering and Electronics,2022,44(4):1301-1309.
24	卞鸿巍, 李安, 王荣颖, 等.导航概论[M].北京: 科学出版社,2023.
	BIAN H W , LI A , WANG R Y , et al.Introduction to Navigation[M].Beijing: Science Press,2023.
25	WANG J H , HUANG Z H .The recent technological development of intelligent mining in China[J].Engineering,2017,3(4):439-444. doi: 10.1016/J.ENG.2017.04.003
26	CHEN L , MA Y F .Shamanskii-like Levenberg-Marquardt method with a new line search for systems of nonlinear equations[J].Journal of Systems Science and Complex,2020,33(1):1694-1707.

神经元个数	INS独立/m			神经网络修正后/m
神经元个数	均值	均方根	最大值	均值	均方根	最大值
30	1 732.6	329.4	2 285.4	980.2	404.8	2 299.3
40	1 732.6	329.4	2 285.4	1 307.3	584.8	2 877.0
50	1 732.6	329.4	2 285.4	640.9	342.4	1 152.8
60	1 732.6	329.4	2 285.4	921.6	355.4	2 078.9

模型	迭代周数	训练时间/s	占用内存/MB	模型计算时间/s	误差均值/m	误差均方根/m	误差最大值/m
L-M	18	164	125.452	0.35	2 941.2	2 190.1	16 803.2
Bayesian	14	128	68.518	0.33	640.9	342.4	1 152.8

训练样本分配	误差均值/m	误差均方根/m	误差最大值/m
方案A	1 283.0	574.4	2 798.7
方案B	640.9	342.4	1 152.8
方案C	709.3	357.9	1 964.2

[1]	冷悦, 钟胜. 天文/惯性组合系统中重力扰动补偿方法[J]. 系统工程与电子技术, 2024, 46(4): 1357-1363.
[2]	李新宇, 周召发, 张志利, 郝诗文, 梁哲. 基于SINS/OD失准角观测的垂线偏差测量方法[J]. 系统工程与电子技术, 2024, 46(3): 1067-1074.
[3]	郭士荦, 王春雨, 李洋, 田鹏. 卡尔曼滤波器异常自检及其在SINS初始对准中的应用[J]. 系统工程与电子技术, 2024, 46(2): 668-674.
[4]	胡瑞贤, 张昭, 骆成. 一种分布式卫星系统基线优化设计方法[J]. 系统工程与电子技术, 2023, 45(8): 2423-2437.
[5]	查峰, 位秋硕, 何泓洋, 李豹. 基于IMU体对角线旋转的双轴旋转方案[J]. 系统工程与电子技术, 2023, 45(8): 2546-2554.
[6]	薛海建, 王涛, 蔡星会, 王金涛, 江英. 里程计辅助的车载SINS行进间对准方法[J]. 系统工程与电子技术, 2023, 45(6): 1805-1813.
[7]	沈子涵, 赵修斌, 张闯, 张良, 刘鑫贤. 基于长短期记忆神经网络的自适应容错方法[J]. 系统工程与电子技术, 2023, 45(3): 831-838.
[8]	靳凯迪, 柴洪洲, 宿楚涵, 向民志, 李明. 基于状态变换卡尔曼滤波的DVL/SINS组合导航算法[J]. 系统工程与电子技术, 2023, 45(11): 3624-3631.
[9]	吴宗收, 汪立新, 沈强, 李灿, 李文华. 基于CEEMD半球谐振陀螺输出预测方法[J]. 系统工程与电子技术, 2023, 45(10): 3259-3264.
[10]	徐庚, 何永旭, 张勇刚. 基于罗德里格斯参数的惯性系传递对准算法[J]. 系统工程与电子技术, 2022, 44(9): 2903-2913.
[11]	石宸睿, 田露, 徐湛, 职如昕, 陈晋辉. 基于PSO-BP的应急通信感知装备效能评价方法[J]. 系统工程与电子技术, 2022, 44(11): 3455-3462.
[12]	苏瑜, 荆文芳, 卢晓春, 张阳. 基于BP-PID控制的载波频率准确度提高算法[J]. 系统工程与电子技术, 2021, 43(7): 1894-1903.
[13]	王力, 刘子奇. WPA-IGA-BP神经网络的模拟电路故障诊断[J]. 系统工程与电子技术, 2021, 43(4): 1133-1143.
[14]	郝诗文, 张志利, 周召发, 常振军, 刘先一. 重力扰动对惯性导航系统初始对准的影响[J]. 系统工程与电子技术, 2020, 42(7): 1575-1581.
[15]	弋佳东, 杨洁. 基于IFOA-SA-BP神经网络的雷达信号识别[J]. 系统工程与电子技术, 2020, 42(12): 2735-2741.