Font Size: a A A

Control Methods For Upper Limb Rehabilitation Robot Based On Surface Electromyography

Posted on:2016-04-27Degree:MasterType:Thesis
Country:ChinaCandidate:D S LiuFull Text:PDF
GTID:2428330542457485Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
With the social development,the change of environment,and the modern competition pressure,people's diet structure has been unhealthy.It leads to the incidence of stroke,which was increasing year by year.According to the result of epidemiological surveys,about 70%of the patients survived from the stroke have different degrees of hemiplegia.At present,the treatment of the hemiplegic patients is mainly relied on assistant movements by rehabilitation therapists.During the rehabilitation,one rehabilitation therapist just can help one patient that leads to high cost of treatment and lack of rehabilitation therapists.Hence,it is of great value in hemiplegic patient's rehabilitation to study the motion auxiliary medical robotic technology.Rehabilitation robot mainly achieves passive training by applying the impedance in the opposite direction of the motion to enhance patient rehabilitation in previous studies.However,from the rehabilitation effect,active rehabilitation of patients with rehabilitation training stimulating the initiative will help the patient in the reconstruction and rehabilitation of motor function.Surface electromyography(sEMG)is the body's biological signals generated by muscle contraction,to a certain extent;it can reflect the neuromuscular activity.Patients actively participate in rehabilitation has good effect on rehabilitation.Hence,patients can carry out rehabilitation according to their wishes by adding surface electromyography into rehabilitation robot system.According to the characteristics of stroke patients with unilateral limb function is impaired,this study establishes the system of rehabilitation robot,studies the patients'active control method of upper rehabilitation robot based on surface EMG,achieves the extraction of patients' motion intents,and based on the his own intentions,the patient took rehabilitation trainings.The main contents are as follows:Firstly,this thesis elaborates the generation mechanism and characteristics of electromyography,describes the surface EMG and joint angle acquisition system,collects of eight-channel surface EMG from six muscles and extracts five kinds of electromyography features by time-domain analysis.Secondly,based on surface EMG,the joint quantitative estimation method is studied in this thesis.Taking the extracted features of electromyography as input,and joint angle as the output,a quantitative relationship model between joint angles and surface EMG is established based on support vector machine(SVM)and extreme learning machine(ELM).Gravitational search algorithm(GSA)and improved gravitational search algorithm(IGSA)are used to find the optimal parameters for SVM model in order to get a better predictive effect.Thirdly,the kinematics model of the laboratory rehabilitation manipulator is established according to the D-H method.This thesis describes the basic principles of the impedance control especially position-based impedance control,and demonstrates the impedance control feasibility by simulation experiments.After that,this thesis builds experimental system based on 3-DOF rehabilitation robot in the laboratory,completes a surface EMG based rehabilitation robot control experiment with normal human as experimental subject,and verifies the feasibility of rehabilitation system based on surface EMG.Finally,the thesis summarizes the main works,points out its drawback and the future work.
Keywords/Search Tags:surface EMG, support vector machine, gravitational search algorithm, rehabilitation robot, impedance control
PDF Full Text Request
Related items