Int J Performability Eng ›› 2018, Vol. 14 ›› Issue (9): 2175-2180.doi: 10.23940/ijpe.18.09.p26.21752180

Previous Articles     Next Articles

Adaptive Classifier based on Distance of Probabilistic Fuzzy Set for EMG Robot

Wenjing Huanga, Yaoqing Renb, *, Kejun Lia, and Yihua Lic   

  1. aSchool of Materials Science and Engineering, Central South University of Forestry and Technology, Changsha, 410004, China;
    bSchool of Mathematics and Physics, Yancheng Institute of Technology, Yancheng, 224051, China;
    cSchool of Logistics and Traffic, Central South University of Forestry and Technology, Changsha, 410004, China
  • Revised on ; Accepted on
  • Contact: * E-mail address: yqren@aliyun.com

Abstract: Surface electromyographic (sEMG) signals always change with the external and internal conditions of human beings. Such a time-varying characteristic leads to decreasing classification accuracy of fixed-parameter classifiers for EMG patterns with time. To design a control system for EMG-based artificial limbs with stable performance, it is necessary to introduce the adaptive mechanism in the classifiers for EMG patterns. In addition, there are many uncertainties in the process of EMG signal acquisition and grasp model recognition. In this paper, on the basis of a distance classifier based on probabilistic fuzzy set, we attempted to introduce the adaptive scheme to the classifiers for EMG patterns and then verified the application of the scheme in the classification of EMG patterns through experiments. The study shows that a self-enhancement distance classifier based on probabilistic fuzzy set can improve recognition accuracy.

Key words: adaptive classification, probabilistic fuzzy, electromyographic robot