| In recent years,as China's aging population has become increasingly serious,the number of stroke patients has also increased year by year.About 75% of survivors after stroke have varying degrees of disability,and they often have sequelae such as muscle stiffness and loss of motor function.At present,the main method of rehabilitation is to use rehabilitation doctors to perform a series of high-intensity rehabilitation training with hands-on assisted patients in a one-to-one manner.This type of rehabilitation mainly involves problems such as large amounts of labor,high costs,and limited time and place.For this type of problem,a master-slave type finger rehabilitation robot is designed,which is mainly used for assisting hemiplegia patients in hand rehabilitation exercises.By recognizing the hand movement of the patient's healthy limb and using the recognition result to control the manipulator worn by the limb,a two-handed tracking rehabilitation training is realized.In this paper,the electromyographic(EMG)signals of the arms and the gesture signals of the fingers are collected,and the original signals are detected by active segment detection and feature extraction.The eigenvalue replaces the original signal as the input of a Support Vector Machine(SVM)classification model to achieve gesture recognition.In order to further improve the accuracy of the classification model,the particle swarm optimization algorithm is used to optimize the parameters of the SVM classification model.Experimental results show that the correct rate of gesture recognition is as high as 99.6%.In this paper,a three-finger exoskeleton-type manipulator is designed based on the hand movement characteristics and structural characteristics.It is used to assist finger joints to complete certain strength bending and stretching exercises.Using Solidworks to complete the three-dimensional modeling of the manipulator,and establishing a kinematic model of the knuckle joint,performing kinematics simulation analysis to verify the rationality of the structural design.After receiving the control command,the STM32 controller realizes the linkage of multiple joints of the manipulator by driving a plurality of motors at the same time,thereby pulling the fingers of the affected limb to complete similar actions with the fingers of the health limb and achieving a certain degree of gesture tracking.Finally,the finger rehabilitation training simulation experiment proves that each module of this tracking finger rehabilitation system has good function,reasonable mechanical structure design,and basically achieves the desired goal. |
