基于表面肌电信号估计多自由度手指力的ShuffleVT模型  

ShuffleVT model for estimating multi-degree-of-freedom finger forcebased on surface electromyography

作　　者：李思诺 耿艳娟[2] Li Sinuo;Geng Yanjuan(College of Mechanical and Control Engineering,Guilin University of Technology,Guilin 541006,China;Shenzhen Institute of Advanced Technology,Chinese Academy of Sciences,Shenzhen 518055,China)

机构地区：[1]桂林理工大学机械与控制工程学院,桂林541006 [2]中国科学院深圳先进技术研究院,深圳518055

出　　处：《电子测量技术》2025年第5期102-110,共9页Electronic Measurement Technology

基　　金：国家自然科学基金(62373345,82161160341);深圳市基础研究重点项目(JCYJ20220818101602005)资助。

摘　　要：为了实现基于表面肌电信号(sEMG)的连续手指力估计,本文提出了一种融合ShuffleNetV2基本单元与Vision Transformer(ViT)结构的新模型,命名为ShuffleVT。为验证该模型的性能,采用公开数据集NinaPro,其包含40名健康受试者的sEMG数据和6个自由度的手指力数据。性能评估指标为Pearson相关系数(CC)、均方根误差(RMSE)和决定系数(R^(2))。结果显示,ShuffleVT模型的CC、RMSE和R^(2)平均值分别为0.92±0.05、1.27±0.66和0.83±0.10,显著优于ShuffleNetV2、ViT、Transformer和LSTM等4种深度学习模型。该结果展示了ShuffleVT模型在基于表面肌电的连续运动意图估计中的应用潜力。In this study,a ovel deep learning model named ShuffleVT was proposed to estimate continuous finger force base on surface electromyography(sEMG).This model was composed of the basic units of ShuffleNetV2 and the Vision Transformer(ViT)structure.Its performance was validated on a publicly available NinaPro database,which includes sEMG data from 40 healthy subjects and finger force data from 6 degrees of freedom.Three performance metrics including Pearson correlation coefficient(CC),root mean square error(RMSE),and coefficient of determination(R^(2))were used.And another four deep learning models,ShuffleNetV2,ViT,Transformer,and LSTM were included for comparison.The results showed the averaged CC,RMSE,and R^(2) was 0.92±0.05,1.27±0.66,and 0.83±0.10,respectively,significantly better than that computed with another four models.It indicates that the newly proposed ShuffleVT model could by potentially applied into the sEMG-driven continuous estimation of human motor intention.

关 键 词：表面肌电信号 手指力 连续估计 深度学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN911.7[自动化与计算机技术—计算机科学与技术]