作者: Guoyuan Tang , Haoping Wang , Yang Tian
关键词:
摘要: The research of surface Electromyography (sEMG) has developed rapidly in recent years, mainly the recognition motion intention and prediction joint angles. Consequently, sEMG come to foreground as an ideal human-computer interface. This paper proposes a method which applies sliding window based on energy value first differential signal (DSWE) achieve more stable effect for determination onset time. A combination 4 time domain features 19 fuzzy entropy wavelet subspaces are selected recognition. Afterwards, classification estimation model were established utilizing BP neural network. To verify effectiveness network, three patterns including standing up, flexion extension conducted respectively. result presents that average accuracy human is 98.3% 3-channel system can estimate knee angles with 3.25-11.65% error.