Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

A robot scheduling method based on rMAPPO for H-beam riveting and welding work cell.

PloS one·2025
Same author

The Effects of Warfarin on the Pharmacokinetics of Senkyunolide I in a Rat Model of Biliary Drainage After Administration of Chuanxiong.

Frontiers in pharmacology·2019
Same author

Biomarker Discovery for Immunotherapy of Pituitary Adenomas: Enhanced Robustness and Prediction Ability by Modern Computational Tools.

International journal of molecular sciences·2019
Same author

5-Aminothiophene-2,4-dicarboxamide analogues as hepatitis B virus capsid assembly effectors.

European journal of medicinal chemistry·2018
Same author

Novel Hepatitis B Virus Capsid-Targeting Antiviral That Aggregates Core Particles and Inhibits Nuclear Entry of Viral Cores.

ACS infectious diseases·2018
Same author

Gene expression profiling reveals differential patterns between microcystic congenital cystic adenomatoid malformation and congenital lobar emphysema.

Early human development·2018

相关实验视频

Updated: Jan 17, 2026

Training Persons with Spinal Cord Injury to Ambulate Using a Powered Exoskeleton
09:46

Training Persons with Spinal Cord Injury to Ambulate Using a Powered Exoskeleton

Published on: June 16, 2016

21.3K

基于对外骨在不同地形上的转移学习的模式过渡识别.

Yifan Gao1, Jianbin Zheng1, Yang Gao1

  • 1School of Information Engineering, Wuhan University of Technology, Wuhan, Hubei, China.

PeerJ. Computer science
|September 24, 2025
PubMed
概括

这项研究引入了一种新的转移学习方法,用于在可穿戴机器人中检测人类运动意图. 该方法准确地预测了移动模式的转换,使外骨在不同地形上更顺地导航.

关键词:
模式转换识别模式的识别.在T-T前期.在T前/GC前.在 TCN-SA 中.转移学习转移学习

更多相关视频

Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis
11:16

Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis

Published on: July 22, 2014

16.7K
A Rehabilitation Program of Exoskeleton-assisted Body Weight-Supported Treadmill Training with Non-immersive Virtual Reality for Stroke Patients
06:00

A Rehabilitation Program of Exoskeleton-assisted Body Weight-Supported Treadmill Training with Non-immersive Virtual Reality for Stroke Patients

Published on: May 16, 2025

999

相关实验视频

Last Updated: Jan 17, 2026

Training Persons with Spinal Cord Injury to Ambulate Using a Powered Exoskeleton
09:46

Training Persons with Spinal Cord Injury to Ambulate Using a Powered Exoskeleton

Published on: June 16, 2016

21.3K
Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis
11:16

Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis

Published on: July 22, 2014

16.7K
A Rehabilitation Program of Exoskeleton-assisted Body Weight-Supported Treadmill Training with Non-immersive Virtual Reality for Stroke Patients
06:00

A Rehabilitation Program of Exoskeleton-assisted Body Weight-Supported Treadmill Training with Non-immersive Virtual Reality for Stroke Patients

Published on: May 16, 2025

999

科学领域:

  • 机器人和人机交互的人机交互
  • 机器学习和模式识别.

背景情况:

  • 人类运动意图检测对于先进的可穿戴机器人至关重要.
  • 识别运动模式 (如步行,爬楼梯) 之间的过渡是具有挑战性的,特别是在不同的物理负载和地形下.

研究的目的:

  • 开发和评估一种新的转移学习方法,用于准确和早期识别移动模式模式转换.
  • 提高可穿戴机器人在适应不同地形和物理条件方面的性能.

主要方法:

  • 一种结合时间卷积网络 (TCN) 与空间注意力 (SA) 的转移学习方法被开发用于模式过渡识别.
  • 该方法在三重物理负载下在五种动态模式中识别八种运动模式之间的过渡时进行了测试.
  • 与ResNet和LSTM等其他模型相比,基于准确性和预测时间 (Pre-T) 评估了性能.

主要成果:

  • 在TCN-空间注意力 (TCN-SA) 转移学习方法实现了高精度,达到高达98.21%.
  • 实现了对下一个移动模式的早期预测,预测时间在实际过渡之前从120-600 ms不等.
  • 步行周期 (Pre-T/GC) 中预测时间的比例显著降低,表明有效的预测.

结论:

  • 提议的TCN-SA转移学习方法有效地检测出运动模式的转换,从而能够更早地进行预测.
  • 这种早期检测允许可穿戴机器人,如外骨,在相邻的地形之间更平稳,更有效地导航.
  • 该方法在预测时间和准确性方面表现出比现有方法更好的性能.