曹尔凡,崔建伟.基于数据手套与虚拟模型的手臂动作识别方法[J].测控技术,2021,40(1):16-21
基于数据手套与虚拟模型的手臂动作识别方法
Arm Motion Recognition Method Based on Data Gloves and Virtual Model
  
DOI:10.19708/j.ckjs.2020.11.324
中文关键词:  数据手套  手臂模型  动作轨迹特征  动作识别  支持向量机
英文关键词:data gloves  arm model  motion track characteristics  motion recognition  SVM
基金项目:国家自然科学基金项目(61873063)
作者单位
曹尔凡 东南大学 仪器科学与工程学院 
崔建伟 东南大学 仪器科学与工程学院 
摘要点击次数: 473
全文下载次数: 219
中文摘要:
      为了配合残疾人使用助残假手完成日常活动,提出了一种基于数据手套与虚拟模型的手臂动作识别方法。通过数据手套上的MEMS传感器测量手臂运动数据,结合虚拟交互环境中的手臂模型,完成模型对真实手臂的动作复现,并获取手臂模型的位置数据。而后将传感器加速度、角速度等数据的时域特征与手臂模型位置数据的动作轨迹特征结合起来,使用支持向量机算法对预定义的5种手臂动作进行分类识别,识别率达到99.33%,验证了提出的手臂动作识别方法的可行性。
英文摘要:
      In order to cooperate with disabled people to use hand-assisted prosthetic hands to complete daily activities,a method of arm motion recognition based on data gloves and virtual models is proposed.The arm motion data was measured by the MEMS sensors on the data glove,combined with the arm model in the virtual interactive environment,the motion of the model to the real arm was completed,and the position data of the arm model was obtained.Then,the time domain characteristics of sensor acceleration,angular velocity and other data were combined with the motion trajectory characteristics of the arm model position data,and the support vector machine(SVM) algorithm was used to classify and recognize the five predefined arm movements.The recognition rate reached 99.33%.It verifies the feasibility of the proposed arm motion recognition method.
查看全文  查看/发表评论  下载PDF阅读器
关闭