Research On 4-DOF Cable-Driven Upper Limb Rehabilitation Exoskeleton Robot Technology

Stroke is a severe disease caused by rupture or obstruction of blood vessels in the brain,resulting in brain tissue damage.It has the characteristics of high morbidity,high mortality and high disability rate.It usually leads to hemiplegia/paralysis in the limbs of patients,thus limiting their ability to daily life.With the aging of society and the acceleration of urbanization,the incidence of stroke increases rapidly,and stroke has become a major threat to human beings.When the brain is damaged by stroke,it has the ability to reshape itself,known as neuroplasticity.High-insensitive and repetitive oriented exercise training has a positive impact on the recovery of limb motor function.However,due to the limited number of therapists and their work intensity,rehabilitation training is usually difficult to carry out.At present,it is of great significance to carry out rehabilitation training for stroke patients by robots,which can reduce the pressure of therapists,improve the efficiency of rehabilitation training,reduce the cost of rehabilitation training,and realize real-time monitoring of patients’ motor ability and physiological information through a variety of sensors of robot devices,so as to formulate reasonable training strategies.Based on the above background,aiming at the rehabilitation training requirements of upper limb hemiplegia caused by stroke,a cable-driven 4-DOF upper limb rehabilitation exoskeleton robot was designed.The main contents of the paper are as follows:Based on the structural characteristics of human upper limbs and the related needs of stroke patients in rehabilitation training,the problems of small effective workspace and low cable tension efficiency caused by the unidirectionality of cable tension were investigated.A dynamic adaptive structure of cable attachment points was proposed,and a 4-DOF cable-driven exoskeleton based on this structure was designed for shoulder flexion/extension,adduction/abduction,internal/external rotation and elbow flexion/extension rehabilitation training,including the design of the exoskeleton structure and the design of the sensing and control system.The joint coordinate system and kinematics model of the exoskeleton robot were established.The initial distribution of cable attachment points of the cable-driven exoskeleton with dynamic adaptive structure was optimized.The effective workspace and cable tension efficiency of the exoskeleton were analyzed and compared with the existing cable-driven exoskeleton.The results show that the dynamic adaptive structure proposed in this paper can effectively improve the workspace and cable tension efficiency of the cable-driven exoskeleton.The driving characteristics of cable-driven exoskeleton system were analyzed.A magnetorheological(MR)actuator with low inertia/high torque and an active/semi-active hybrid drive system based on MR actuators have been designed.Compared with the traditional driven system based on electric motor actuators,the hybrid driven system was more in line with the cable-driven system of antagonistic,unidirectional and dynamic performance requirements.Moreover,it had certain advantages in performance,volume,quality and cost.The performance of cable-driven system based on active/semi-active hybrid driven system was verified by experiments.Aiming at the structural characteristics of the cable-driven upper limb rehabilitation exoskeleton,a parameter identification algorithm was designed to identify some parameters such as the patient’s limb length,the wearing position and the center shift of shoulder joint,which reduced the inconvenience and error caused by manual measurement.A self-tensioning algorithm of the cable-driven exoskeleton was designed to eliminate the limb ring displacement along the arm direction caused by the redundant force in the working process of the cabledriven system,so as to improve the discomfort caused by the over-tight binding of the traditional cable-driven exoskeleton arm ring and reduce the systematic error caused by the arm ring displacement.Aiming at the passive training needs of patients with hemiplegia in acute stage of stroke,the limb movement of patients with exoskeleton traction along the desired trajectory was realized based on mapping PID control algorithm.The assist-as-needed training algorithm was designed for patients in the convalescent stage,and the robot-assisted mode was adjusted according to the trajectory tracking error of patients performing active training tasks,so as to promote patients to actively participate in training tasks as much as possible and improve the efficiency of rehabilitation training.Dynamic adaptive structure controllers for passive and active training were designed respectively.Aiming at the proposed cable and active/semi-active hybrid driven system,the safety characteristics were analyzed,and three working modes were proposed to improve the function of the upper limb rehabilitation training robot.Experiments were carried out to verify the effectiveness of the control algorithm and working mode.