Evaluation of Muscle Fatigue Based on CRP and RQA for Upper Limb Exoskeleton

摘要

The movement of lifting heavy objects with upper limb exoskeleton is a dynamic process of multi-muscle coordination. Muscle fatigue occurs after repetitive lifting heavy objects with exoskeleton robot for a long time. Due to the dynamic characteristics of the moving process, the traditional linear analysis methods can not fully characterize the muscle fatigue problem of upper limb. In this paper, a novel approach base on ross recurrence plot (CRP)and recurrence quantification analysis (RQA)is proposed to analyze the progression of muscle fatigue of upper limb. Firstly, distance threshold and date length of the recurrence plot is chosen according to the RP dynamical structures of the electromyography (EMG)signals. Then the phase space reconstruction techniques and recurrence plot theory are utilized to analyze the nonstationarity characteristics of the multichannel and single-channel surface EMG signals during the fatigue process. Finally, the RQA indices of the overall and local muscles are calculated. The experiments show that the CRP can qualitatively reflect the change of muscle fatigue state. Meanwhile, the RQA indices of multi-channel and single-channel EMG signals can quantitatively reveal the changes of muscle fatigue from the overall and local perspectives. This method provides a theoretical basis for evaluating muscle fatigue during repeated moving in the way of nonlinear dynamics.