China Mechanical Engineering ›› 2022, Vol. 33 ›› Issue (17): 2071-2078.DOI: 10.3969/j.issn.1004-132X.2022.17.008

Previous Articles     Next Articles

Intention Recognition Method of Human Robot Cooperative Assembly Based on EEG-EMG Signals

DONG Yuanfa1,2;JIANG Lei2;PENG Wei1,2;ZHOU Bin1,2;FANG Zifan2   

  1. 1.Intelligent Manufacturing Innovation Technology Centre,China Three Gorges University,Yichang,Hubei,443002
    2.College of Mechanical & Power Engineering,China Three Gorges University,Yichang,Hubei,443002
  • Online:2022-09-10 Published:2022-09-23

融合脑电-肌电信号的人机协作装配意图识别方法

董元发1,2;蒋磊2;彭巍1,2;周彬1,2;方子帆2   

  1. 1.三峡大学智能制造创新技术中心,宜昌,443002
    2.三峡大学机械与动力学院,宜昌,443002
  • 通讯作者: 彭巍(通信作者),男,1986年生,讲师。研究方向为人机共融与智能制造等。E-mail:pengwei_tju@163.com。
  • 作者简介:董元发,男,1988年生,博士、副教授。研究方向为MBSE、人机共融和智能制造。E-mail:d828891@163.com。
  • 基金资助:
    国家自然科学基金(52075292);湖北省自然科学基金(2019CFB542);水电机械设备设计与维护湖北省重点实验室开放基金(2020KJX05)

Abstract:  Aiming at the problems of low accuracy and poor stability of identifying cooperative intention based on single source physiological electrical signals in human robot cooperative assembly scene, firstly, support vector machine was used to recognize single source cooperative assembly intentions from EEG and EMG signals respectively, and then D-S evidence theory was used to fuse the recognition results of multi-source cooperative assembly intentions. A human robot collaborative assembly intention recognition method was proposed based on EEG-EMG signals. Experimental results show that the proposed method may effectively improve the accuracy and stability of human robot cooperative assembly intention recognition. 

Key words: human robot collaboration, assembly, intention recognition, electroencephalogram(EEG) signals, electromyogram(EMG) signals

摘要: 针对人机协作装配场景下基于单源生理电信号识别协作意图准确率不高、稳定性不好的问题,首先采用支持向量机方法分别从脑电信号和肌电信号识别单源协作装配意图,然后采用D-S证据理论对多源协作装配意图识别结果进行融合,提出了一种融合EEG-EMG信号的人机协作装配意图识别方法。实验结果表明,所提方法可以有效提高人机协作装配意图识别的准确率和稳定性。

关键词: 人机协作, 装配, 意图识别, 脑电信号, 肌电信号

CLC Number: