Research On Robot-assisted Training Mode Of Human-machine Interactive End-effector Robot

With the development of my china’s economy and the acceleration of the aging process of the population,the demand for driving cars by the elderly is also increasing.Cerebral hemorrhage caused by car accidents and stroke have severely damaged the motor function of patients.Rehabilitation training can effectively treat hemiplegia caused by cerebral hemorrhage of car accidents and stroke.The traditional rehabilitation treatment method is that therapist guide the patient to perform limb training,which is not only labor-intensive and time-consuming,but also the training mode is simple.Compared with the traditional manual mode,robot-assisted rehabilitation training has the advantages of high accuracy,quantification,strong reproducibility and abundant training modes,which can greatly improve the efficiency of rehabilitation training.In clinical practice,the assessment of patients’ motor function is separated from training,so it is impossible to judge the movement state of the subjects in training in real time.Based on the end-effector upper limb rehabilitation robot,this paper aims to combine the robot-assisted training technology of rehabilitation robot with the motor function assessment method of patients,so as to provide an efficient rehabilitation training method for patients,which is of great significance to make up for the deficiency of traditional rehabilitation training and improve the efficiency of rehabilitation training.Firstly,according to the clinical rehabilitation mechanism of hemiplegia after stroke,the research on the overall system scheme of the plane end-effector upper limb rehabilitation robot was mainly carried out,and its mechanical structure,hardware system and software system were analyzed and designed respectively.Secondly,the paper takes the end-effector upper limb rehabilitation robot as the research object,and designs the robot-assisted training mode based on the admittance control theory.The patient’s motion intention is extracted by the force sensor at the end of the robot.Based on the position control,the admittance control is used as the outer loop control,and the interactive force between the human and the robot is converted into position information to realize a compliant interactive control.The simulation model was built by Matlab and combined with clinical experiments to analyze the influence of different admittance parameters on patients’ training performance,which provided a reliable basis for parameter selection in rehabilitation training and also verified the feasibility of the algorithm in the actual rehabilitation training system.Finally,the paper combines the patient’s motor function assessment with the upper limb rehabilitation robot assisted training mode,and establishes a method for evaluating the motor function of hemiplegic patients based on the movement trajectory and interaction force information in the robot-assisted training mode.The circular trajectory training involves the coordinated movement of the upper limbs’ multi-pack muscle groups and is used as a method to evaluate the patient’s movement function based on the robot training technology.The characteristic parameters such as movement deviation,movement velocity and movement smoothness are calculated by the movement trajectory and interaction force information.The paper analyzes the differences between the characteristic parameters of hemiplegic patients and healthy subjects in detail,and the parameter set of motor functional quantitative assessment of patients with hemiplegia was extracted based on movement trajectory and interaction force information.Through multi-linear regression analysis of quantitative parameters and the clinical upper limb Fugl-Meyer score,a method of motor functional assessment for hemiplegia patients was established based on robot-assisted training mode.