Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments

Performing long-duration navigation without the global navigation satellite system (GNSS) network is a challenging task, particularly for small unmanned aerial vehicles (UAVs) equipped with low-cost micro-electro-mechanical sensors. This study proposes a hybrid neural network that integrates self-at...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziyi Wang, Xiaojun Shen, Jie Li, Juan Li, Xueyong Wu, Yu Yang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Drones
Subjects:
Online Access:https://www.mdpi.com/2504-446X/9/4/279
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850144992900677632
author Ziyi Wang
Xiaojun Shen
Jie Li
Juan Li
Xueyong Wu
Yu Yang
author_facet Ziyi Wang
Xiaojun Shen
Jie Li
Juan Li
Xueyong Wu
Yu Yang
author_sort Ziyi Wang
collection DOAJ
description Performing long-duration navigation without the global navigation satellite system (GNSS) network is a challenging task, particularly for small unmanned aerial vehicles (UAVs) equipped with low-cost micro-electro-mechanical sensors. This study proposes a hybrid neural network that integrates self-attention mechanisms with long short-term memory (SALSTM) to enhance GNSS-denied navigation performance. The estimation task of GNSS-denied navigation is first modeled based on UAV aerodynamics and kinematics, enabling a precise definition of the inputs and outputs that SALSTM needs to map. A self-attention layer is inserted in multiple LSTM layers to capture long-range dependencies in subtle dynamic changes. The output layer is designed to generate state sequences, leveraging the recursive nature of LSTM to enforce state continuity constraints. The outputs of SALSTM are fused to enhance integrated navigation within an extended Kalman filter framework. The performance of the proposed method is evaluated using flight data obtained from field tests. The results demonstrate that SALSTM-enhanced integrated navigation achieves superior long-term stability and improves velocity and position estimation accuracy by more than 50% compared to the best existing methods.
format Article
id doaj-art-e7dcba2ad16947e78fd957789e823f2d
institution OA Journals
issn 2504-446X
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Drones
spelling doaj-art-e7dcba2ad16947e78fd957789e823f2d2025-08-20T02:28:12ZengMDPI AGDrones2504-446X2025-04-019427910.3390/drones9040279Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied EnvironmentsZiyi Wang0Xiaojun Shen1Jie Li2Juan Li3Xueyong Wu4Yu Yang5School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, ChinaBeijing Special Electromechanical Research Institute, Beijing 100012, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, ChinaYangtze Delta Region Academy of Beijing Institute of Technology, No. 1940, East North Road, Youchegang Town, Xiuzhou District, Jiaxing 314019, ChinaPerforming long-duration navigation without the global navigation satellite system (GNSS) network is a challenging task, particularly for small unmanned aerial vehicles (UAVs) equipped with low-cost micro-electro-mechanical sensors. This study proposes a hybrid neural network that integrates self-attention mechanisms with long short-term memory (SALSTM) to enhance GNSS-denied navigation performance. The estimation task of GNSS-denied navigation is first modeled based on UAV aerodynamics and kinematics, enabling a precise definition of the inputs and outputs that SALSTM needs to map. A self-attention layer is inserted in multiple LSTM layers to capture long-range dependencies in subtle dynamic changes. The output layer is designed to generate state sequences, leveraging the recursive nature of LSTM to enforce state continuity constraints. The outputs of SALSTM are fused to enhance integrated navigation within an extended Kalman filter framework. The performance of the proposed method is evaluated using flight data obtained from field tests. The results demonstrate that SALSTM-enhanced integrated navigation achieves superior long-term stability and improves velocity and position estimation accuracy by more than 50% compared to the best existing methods.https://www.mdpi.com/2504-446X/9/4/279fixed-wing UAVintegrated navigationGNSS-deniedself-attentionLSTM
spellingShingle Ziyi Wang
Xiaojun Shen
Jie Li
Juan Li
Xueyong Wu
Yu Yang
Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments
Drones
fixed-wing UAV
integrated navigation
GNSS-denied
self-attention
LSTM
title Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments
title_full Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments
title_fullStr Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments
title_full_unstemmed Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments
title_short Enhancing Integrated Navigation with a Self-Attention LSTM Hybrid Network for UAVs in GNSS-Denied Environments
title_sort enhancing integrated navigation with a self attention lstm hybrid network for uavs in gnss denied environments
topic fixed-wing UAV
integrated navigation
GNSS-denied
self-attention
LSTM
url https://www.mdpi.com/2504-446X/9/4/279
work_keys_str_mv AT ziyiwang enhancingintegratednavigationwithaselfattentionlstmhybridnetworkforuavsingnssdeniedenvironments
AT xiaojunshen enhancingintegratednavigationwithaselfattentionlstmhybridnetworkforuavsingnssdeniedenvironments
AT jieli enhancingintegratednavigationwithaselfattentionlstmhybridnetworkforuavsingnssdeniedenvironments
AT juanli enhancingintegratednavigationwithaselfattentionlstmhybridnetworkforuavsingnssdeniedenvironments
AT xueyongwu enhancingintegratednavigationwithaselfattentionlstmhybridnetworkforuavsingnssdeniedenvironments
AT yuyang enhancingintegratednavigationwithaselfattentionlstmhybridnetworkforuavsingnssdeniedenvironments