Joint neural network and diagonal loading-based GNSS outage positioning algorithm
-
摘要: GNSS/INS组合导航系统可以为移动载体提供长时间、高精度的导航信息，然而当载体处于恶劣环境中，无法获得滤波量测向量，导致导航定位结果迅速发散. 为应对这一问题，越来越多的学者利用人工神经网络辅助组合导航系统直接进行信息融合. 但惯性导航系统(inertial navigation system，INS)本身特性使得上一时间训练好的网络模型存在误差，中断时刻INS误差仍不断累积，因此提出了一种GNSS中断时的智能定位算法. 该算法利用反向传播(back propagation，BP)神经网络训练得到滤波量测向量，再通过对角加载重构量测噪声协方差矩阵，对卡尔曼滤波(Kalman filter，KF)进行更新. 所提方法减弱了神经网络训练误差对组合导航算法的影响，从而可以在GNSS信号长时间中断情况下，导航系统仍拥有较为可靠的导航性能.
-
关键词:
- 组合导航 /
- 捷联惯导(SINS) /
- 神经网络 /
- 重构协方差矩阵 /
- 卡尔曼滤波(KF)
Abstract: The GNSS/INS integrated navigation system can provide long-term, high-precision navigation information for mobile carriers. However, in adverse environments where filter measurement vectors cannot be obtained, it leads to rapid divergence in navigation positioning results. To address this issue, an increasing number of researchers are employing artificial neural networks to directly fuse information in the integrated navigation system. However, the inherent characteristics of the inertial navigation system (INS) result in errors in previously trained network models, and inertial navigation errors continue to accumulate during interruption periods. Therefore, an intelligent positioning algorithm for GNSS interruptions is proposed. This algorithm utilizes backpropagation (BP) neural networks to train filter measurement vectors and then updates the Kalman filter (KF) by incorporating diagonal-loaded reconstructed measurement noise covariance matrices. This approach reduces the impact of neural network training errors on the integrated navigation algorithm, enabling the navigation system to maintain relatively reliable navigation performance even during prolonged GNSS signal interruptions. -
表 1 INS误差参数
参数 值 陀螺仪零偏 0.015°/h 陀螺仪随机游走 0.01°/$\sqrt {\mathrm{h}} $ 加速度计常值零偏 100 μg 加速度计随机游走 10 μg/$\sqrt {\mathrm{Hz}} $ 
下载: 导出CSV
表 2 BP神经网络参数设置
参数 值 隐层神经元个数 13 训练函数 trainlm 训练次数 2 000 学习效率 0.001 训练目标最小误差 1×10−8 训练最小梯度 1×10−7
下载: 导出CSV
表 3 不同辅助方法下位置误差统计特性
m 采用的方法 位置误差的均值 位置误差均方根值 东向 北向 天向 东向 北向 天向 传统组合
导航41.77 82.34 85.25 48.34 96.11 100.23 BP网络
代替KF7.43 31.81 38.41 10.23 39.38 47.09 BP-KF 8.31 11.86 18.06 8.32 11.86 24.52 BP-RCMKF 2.75 6.33 2.46 3.33 7.43 7.89
下载: 导出CSV
-
[1] 秦永元. 惯性导航[M]. 北京: 科学出版社, 2006. [2] LEICK A, PAPOPORT L, TATARNIKOV D. GNSS Positioning Approaches [M]. USA: John Wiley & Sons, Inc, 2015. [3] BATUWANGALA E, RAMASAMY S, BOGODA L, et al. An interoperability assessment model for CNS/ATM systems[C]//The 38th Australasian Transport Research Forum, Melbourne, Australia, 2016: 01627442. [4] ISMAIL M, ABDELKAWY E. A hybrid error modeling for MEMS IMU in integrated GPS/INS navigation system[J]. The journal of Global Positioning Systems, 2018, 16(1): 6. DOI: 10.1186/s41445-018-0016-5 [5] ABDEL KAREEM JARADAT M, ABDEL-HAFEZ M F. Non-linear autoregressive delay-dependent INS/GPS navigation system using neural networks[J]. IEEE sensors journal, 2017, 17(4): 1105-1115. DOI: 10.1109/JSEN.2016.2642040 [6] AL BITAR N, GAVRILOV A. A new method for compensating the errors of integrated navigation systems using artificial neural networks[J]. Measurement, 2021, 168(1): 108391. [7] 关翔中, 蔡晨晓, 翟文华, 等. 基于神经网络补偿的室内无人机组合导航系统[J]. 航空学报, 2020, 41(S1): 723790. [8] 王超, 周军, 黄浩乾, 等. BP神经网络辅助的SINS/GPS组合导航姿态误差补偿方法研究[J]. 电子器件, 2021, 44(4): 987-993. [9] 方伟, 江金光, 谢东鹏. 基于MLP神经网络改进组合导航算法[J]. 计算机工程与设计, 2021, 42(1): 65-69. [10] CHEN L, FANG J C. A hybrid prediction method for bridging GPS outages in high-precision POS application[J]. IEEE transactions on instrumentation and measurement, 2014, 63(6): 1656-1665. DOI: 10.1109/TIM.2013.2292277 [11] 白相文, 杨建华, 杨志强. 神经网络辅助的组合导航算法研究[J]. 导航定位学报, 2020, 8(1): 93-98. DOI: 10.3969/j.issn.2095-4999.2020.01.017 [12] ABDOLKARIMI E S, ABAEI G, MOSAVI M R. A wavelet-extreme learning machine for low-cost INS/GPS navigation system in high-speed applications[J]. GPS solutions, 2018, 22(1): 1-13. DOI: 10.1007/s10291-017-0682-x [13] YUE S, CONG L, QIN H L, et al. A robust fusion methodology for MEMS-based land vehicle navigation in GNSS-challenged environments[J]. IEEE access, 2020, 8: 44087-44099. DOI: 10.1109/ACCESS.2020.2977474 [14] 徐博, 李盛新, 王连钊, 等. 一种基于自适应神经模糊推理系统的多AUV协同定位方法[J]. 中国惯性技术学报, 2019, 27(4): 440-447. [15] DAI H F, BIAN H W, WANG R Y, et al. An INS/GNSS integrated navigation in GNSS denied environment using recurrent neural network[J]. Defence technology, 2020, 16(2): 334-340. DOI: 10.1016/j.dt.2019.08.011 [16] WU F, LUO H Y, JIA Hongwei, et al. Predicting the noise covariance with a multitask learning model for Kalman filter-based GNSS/INS integrated navigation[J]. IEEE transactions on instrumentation and measurement, 2021, 70(1): 1-13. [17] WANG G Q, HAN Y, CHEN J, et al. A GNSS/INS integrated navigation algorithm based on Kalman filter[J]. IFAC-PapersOnLine, 2018, 51(17): 232-237. DOI: 10.1016/j.ifacol.2018.08.151 -
点击查看大图
图(7) / 表(3)
计量
- 文章访问数: 89
- HTML全文浏览量: 29
- PDF下载量: 8
- 被引次数: 0
