Design And Research Of Key Parts Of Hand Rehabilitation Robot System For Stroke Patients

Hand rehabilitation of stroke patients is an important and difficult problem in the field of rehabilitation.Hand-assisted rehabilitation using rehabilitation robots is a research hotspot in recent years.However,research on hand rehabilitation is still insufficient compared to other parts of the human body.To promote the further development of research on hand rehabilitation robots,this paper designed a hand rehabilitation robot system for stroke patients.The system featured: it can drive the affected limb to simulate the hand movement of normal people;the output torque was designed according to the muscle characteristics of stroke patients;the brain-computer interface realized the awareness recognition of complex hand movements,and the patient’s rehabilitation level can be evaluated after each rehabilitation exercise.The main content of this article is as follows:(1)The research progress related to hand rehabilitation robots and the domestic and foreign current research status was introduced.On this basis,this paper put forward the research purpose of this topic and summarized the main research content of this paper.(2)The mechanical structure of the rehabilitation robot was designed.The motion trajectory data of fingers in spatial coordinate system was collected by motion capture system,then dimension reduction,fitting and sampling processing were carried out on the data,and two motion patterns of hand rehabilitation robot were designed.The rationality of the design was verified by comparing with the theoretical trajectory range.The inverse kinematics and forward kinematics were simulated to complete the structural optimization and verified that the structure met the requirements.(3)The motion capture system was used to collect the finger kinematics data to drive the model to simulate the changes of muscle strength and activation degree of the normal human model.By modifying the maximum isometric muscle strength parameters of the finger musculoskeletal model,two kinds of finger musculoskeletal models suitable for hemiplegic side of stroke patients were obtained.The dynamics of the two models were simulated to obtain the changes of muscle strength and activation degree.(4)The external load was added to simulate the torque exerted on the fingers by the rehabilitation robot.Different external loads were applied to different models and different motion stages to get the changes of muscle strength and activation degree.By comparing the results of finger muscle strength and activation degree in the free state,the rationality of output torque of hand rehabilitation robot system was verified.(5)Completed the hardware selection and the design of the signal analysis process of the active control system.Designed the experiment used the functional near infrared spectroscopy equipment of the active control system to collect data and the instantaneous phase was extracted for statistical analysis.The results showed that the activation ratio of channels in hemiplegic hand of stroke patients was higher than that in non hemiplegic hand and hand of normal people.In addition,the accuracy of consciousness recognition of BCI function was tested,and the results showed that average accuracy rate of 4-level classification of hand complex motion tasks was 73.84± 22.57%.