A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors

Aiming at the problems of multiple data sources, multiple data modes, high data dimensions, large redundancy, small and unbalanced sample size, and difficulty in obtaining a large number of labeled data required for training, a deep bidirectional gated recurrent unit (DBGRU) electromagnetic behavior...

Full description

Saved in:
Bibliographic Details
Main Authors: PAN Ming, ZHENG Jingsong, LI Jinliang, FANG Long, YANG Yang, ZHAO Shijie
Format: Article
Language:zho
Published: Harbin University of Science and Technology Publications 2024-08-01
Series:Journal of Harbin University of Science and Technology
Subjects:
Online Access:https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2348
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702350252408832
author PAN Ming
ZHENG Jingsong
LI Jinliang
FANG Long
YANG Yang
ZHAO Shijie
author_facet PAN Ming
ZHENG Jingsong
LI Jinliang
FANG Long
YANG Yang
ZHAO Shijie
author_sort PAN Ming
collection DOAJ
description Aiming at the problems of multiple data sources, multiple data modes, high data dimensions, large redundancy, small and unbalanced sample size, and difficulty in obtaining a large number of labeled data required for training, a deep bidirectional gated recurrent unit (DBGRU) electromagnetic behavior intent recognition model is constructed. By integrating the attention mechanism in the Bidirectional Gated Recurrent Unit (BiGRU) , the feature learning ability of the model is improved, and adaptively assign the weight of different air combat feature information. With DBGRU as the backbone network, a few-shot contrastive learning algorithm based on data augmentation is proposed, which uses the Wasserstein Generative Adversarial Network (WGAN) based on Wasserstein distance to enrich the original data, and uses the contrastive learning framework to mine the rich pattern information in the multimodal data to make up for the lack of few-shot data, so as to accurately predict the behavior intention of electromagnetic targets. The experimental simulation results show that the accuracy of the few-shot contrastive learning algorithm based on data augmentation in predicting the behavior intention of few-shot air combat targets is 91. 13% .
format Article
id doaj-art-15b6ee1ebc184efdbc0ab3ccd244f49f
institution DOAJ
issn 1007-2683
language zho
publishDate 2024-08-01
publisher Harbin University of Science and Technology Publications
record_format Article
series Journal of Harbin University of Science and Technology
spelling doaj-art-15b6ee1ebc184efdbc0ab3ccd244f49f2025-08-20T03:17:40ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832024-08-012904505810.15938/j.jhust.2024.04.006A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation BehaviorsPAN Ming0ZHENG Jingsong1LI Jinliang2FANG Long3YANG Yang4ZHAO Shijie5Southwest China Research Institute of Electronic Equipment, Chengdu 610036 , ChinaSouthwest China Research Institute of Electronic Equipment, Chengdu 610036 , ChinaSouthwest China Research Institute of Electronic Equipment, Chengdu 610036 , ChinaSchool of Automation, Northwestern Polytechnic University, Xi ′an 710129 , ChinaSchool of Automation, Northwestern Polytechnic University, Xi ′an 710129 , ChinaSchool of Automation, Northwestern Polytechnic University, Xi ′an 710129 , ChinaAiming at the problems of multiple data sources, multiple data modes, high data dimensions, large redundancy, small and unbalanced sample size, and difficulty in obtaining a large number of labeled data required for training, a deep bidirectional gated recurrent unit (DBGRU) electromagnetic behavior intent recognition model is constructed. By integrating the attention mechanism in the Bidirectional Gated Recurrent Unit (BiGRU) , the feature learning ability of the model is improved, and adaptively assign the weight of different air combat feature information. With DBGRU as the backbone network, a few-shot contrastive learning algorithm based on data augmentation is proposed, which uses the Wasserstein Generative Adversarial Network (WGAN) based on Wasserstein distance to enrich the original data, and uses the contrastive learning framework to mine the rich pattern information in the multimodal data to make up for the lack of few-shot data, so as to accurately predict the behavior intention of electromagnetic targets. The experimental simulation results show that the accuracy of the few-shot contrastive learning algorithm based on data augmentation in predicting the behavior intention of few-shot air combat targets is 91. 13% .https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2348intent recognitionattention mechanismsgated recurrent unitgenerative adversarial networkscontrastive learning
spellingShingle PAN Ming
ZHENG Jingsong
LI Jinliang
FANG Long
YANG Yang
ZHAO Shijie
A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors
Journal of Harbin University of Science and Technology
intent recognition
attention mechanisms
gated recurrent unit
generative adversarial networks
contrastive learning
title A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors
title_full A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors
title_fullStr A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors
title_full_unstemmed A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors
title_short A Few-shot Learning Method for Intent Analysis of Air Combat Confrontation Behaviors
title_sort few shot learning method for intent analysis of air combat confrontation behaviors
topic intent recognition
attention mechanisms
gated recurrent unit
generative adversarial networks
contrastive learning
url https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2348
work_keys_str_mv AT panming afewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT zhengjingsong afewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT lijinliang afewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT fanglong afewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT yangyang afewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT zhaoshijie afewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT panming fewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT zhengjingsong fewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT lijinliang fewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT fanglong fewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT yangyang fewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors
AT zhaoshijie fewshotlearningmethodforintentanalysisofaircombatconfrontationbehaviors