Res-EMSA: Adaptive Adjustment of Innovation Based on Efficient Multihead Self-Attention in GNSS/INS Tightly Integrated Navigation System

doi:10.1109/LGRS.2024.3365148

CSpace

	Res-EMSA: Adaptive Adjustment of Innovation Based on Efficient Multihead Self-Attention in GNSS/INS Tightly Integrated Navigation System
	Xu, Hongfu 1; Luo, Haiyong 2; Wu, Zijian 1; Zhao, Fang 1
	2024
发表期刊	IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
ISSN	1545-598X
卷号	21 页码:5
摘要	Tightly integrated navigation based on an extended Kalman filter (EKF) has become ubiquitous across domains such as autonomous driving. The accuracy of the innovation matrix plays a crucial role in the filtering process. However, the variance in data quality across different satellites due to various errors, as well as the nonlinear errors introduced by the linearization of the measurement equation and the errors resulting from the Gaussian noise assumption, can lead to inaccuracies in the innovation matrix. We address these aforementioned issues and propose an adaptive solution relying on Resnet-efficient multihead self-attention (Res-EMSA) to adjust the innovation matrix. Specifically, the network model extracts the features of the global navigation satellite system (GNSS) and inertial navigation system (INS) data through the residual network, and then the features are fused with different weights. After that, the EMSA network is utilized for feature mapping, and ultimately, the fully connected layer is used to perform weight matrix regression. Experimental results reveal that the Res-EMSA model exhibits a significant enhancement in positioning accuracy, with a 39% increase compared to the traditional EKF model.
关键词	Technological innovation Global navigation satellite system Navigation Feature extraction Adaptation models Measurement uncertainty Kalman filters Deep learning extended Kalman filter (EKF) innovation matrix integrated navigation
DOI	10.1109/LGRS.2024.3365148
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program
WOS研究方向	Geochemistry & Geophysics ; Engineering ; Remote Sensing ; Imaging Science & Photographic Technology
WOS类目	Geochemistry & Geophysics ; Engineering, Electrical & Electronic ; Remote Sensing ; Imaging Science & Photographic Technology
WOS记录号	WOS:001173135800024
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38783
专题	中国科学院计算技术研究所
通讯作者	Luo, Haiyong; Zhao, Fang
作者单位	1.Beijing Univ Posts & Telecommun, Sch Comp Sci, Beijing 100876, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Res Ctr Ubiquitous Comp Syst, Beijing 100086, Peoples R China
推荐引用方式 GB/T 7714	Xu, Hongfu,Luo, Haiyong,Wu, Zijian,et al. Res-EMSA: Adaptive Adjustment of Innovation Based on Efficient Multihead Self-Attention in GNSS/INS Tightly Integrated Navigation System[J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS,2024,21:5.
APA	Xu, Hongfu,Luo, Haiyong,Wu, Zijian,&Zhao, Fang.(2024).Res-EMSA: Adaptive Adjustment of Innovation Based on Efficient Multihead Self-Attention in GNSS/INS Tightly Integrated Navigation System.IEEE GEOSCIENCE AND REMOTE SENSING LETTERS,21,5.
MLA	Xu, Hongfu,et al."Res-EMSA: Adaptive Adjustment of Innovation Based on Efficient Multihead Self-Attention in GNSS/INS Tightly Integrated Navigation System".IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 21(2024):5.