Research On Key Technologies Of Flexible Wearable Hand Rehabilitation Robo

With the deepening of the aging of the population,the number of stroke patients in China is increasing year by year.Stroke is an acute cerebrovascular disease with a high mortality rate.Hand motor dysfunction is a typical stroke sequela.The traditional way of rehabilitation is that doctors pull patients’ fingers to do rehabilitation exercise for a long time,but there is a serious shortage of rehabilitation doctors and rehabilitation resources.In recent years,with the development of robots and virtual reality technology,the emergence of rehabilitation robots alleviates the contradiction of the shortage of doctors and provides patients with the possibility of rehabilitation at home.In this paper,a soft wearable hand rehabilitation robot system is designed to help stroke patients with hand injury to carry out rehabilitation training and daily life assistance.The specific work is as follows:First,a soft bidirectional actuator and a soft abduction actuator are designed..Based on thermoplastic polyurethane coated fabric,a soft bidirectional actuator that assists finger flexion and extension is fabricated.In order to achieve better rehabilitation effect,a soft abduction drive is designed and the relationship between its output torque and internal air pressure is deduced.Then the design of the pneumatic control system is completed.The system is mainly composed of a miniature air pump,a high-frequency solenoid valve and a control circuit board.The design of a soft wearable hand rehabilitation robot is completed by connecting the actuator integrated into the glove to the nozzle in the medical box through the trachea.Next,a rehabilitation training system software based on cloud platform is designed.Based on Unity3 D,a passive rehabilitation training software is developed to provide patients with multifunctional rehabilitation training and assistance in activities of daily living.A mirror rehabilitation training system based on Leap Motion sensor is developed,a variety of virtual games with human-computer interaction are designed,and a RLSESN＿PID algorithm is proposed to optimize the position tracking control of the hand rehabilitation robot.Using My SQL database and Java Web framework technology,a cloud platform for storing patient rehabilitation data is designed and developed.Finally,the experiment of hand rehabilitation robot system is carried out.Performance tests of the abduction actuator and robot gloves show that the rehabilitation robot can provide sufficient range of motion as well as output force.In the mirror rehabilitation training system,the superiority of the RLSESN＿PID algorithm in control accuracy and response speed is verified.Ten healthy subjects are recruited for ADL experiments to verify the effect of abduction actuator on the success rate of daily object grasping.Finally,seven stroke patients are included in the rehabilitation robot training for four weeks,which proved the effectiveness of the hand rehabilitation robot system in clinical significance.