Study On The Method Of Upper Limb Motion Intention Recognition And Motor Functional Evaluation During Robot-Assisted Rehabilitation

Stroke is a severe neurologically damaging disease,which causing patients with upper limb motor dysfunction,and rehabilitation training will help patients to recover.Traditional physical therapists have high cost and low efficiency.Robot-assisted rehabilitation training based on patient’s motion intention can effectively promote brain motor nerve remodeling and improve rehabilitation efficiency.At the same time,the physicians need to accurately and objectively assess the patient’s motor function in order to learn about rehabilitation process during robot-assisted training.This study starts from the theory of mirror rehabilitation,considering that s EMG of the upper limbs of the healthy side of stroke patients due to hemiplegia is more effective than the diseased side in reflecting the intention of movement.The healthy side upper limb motion intent recognition model is established by using Support Vector Machine method based on the s EMG signal,and then control the robot to assist the affected upper limb to perform the same motion as the healthy upper limb to achieve the purpose of robot-assisted training.And in this study,a Support Vector Machine was used to establish the upper limb movement intention classification model of each subject,and a comparative analysis was carried out in combination with the classical Linear Discriminant Classification method.The result shows that the model established by the SVM method has good accuracy and stability.At the same time,in order to further objectively and quantitatively evaluate the exercise ability of patients during training,this paper proposes an active participation index estimation method based on work calculation.This method is based on the principle that the total amount of system work done by the same rehabilitation training task is constant.The active participation of the patient in the rehabilitation training process is calculated by measuring the work done of the robot,and its effectiveness is verified through experiments.The result shows that,compared with the EMG detection method,the method based on work calculation can easily and effectively estimate the subject’s active participation.Through the upper limb movement intention recognition model established in this paper,the robot can accurately identify the patient’s upper limb movement intention,so that the patient can get sufficient and effective rehabilitation training.And the proposed method of actively participating in the evaluation of the patient’s motor function can also help the doctor to understand the patient’s rehabilitation process,providing a basis for the patient’s personalized rehabilitation training plan and technical supports for the precision assisted rehabilitation training of the robot.