Research On A 7 Degree Of Freedoms Upper Limb Rehabilitation Robot And Training Trajectory Based On Hybrid Mechanism

Upper limb rehabilitation training is an effective approach for patients with neurological disorders such as stroke to regain their ability for self-care and improve their quality of life.Timely and correct upper limb rehabilitation treatment and training can significantly enhance the rehabilitation outcomes.As China’s population is aging,the incidence of stroke is increasing year by year.In the traditional rehabilitation process,rehabilitation therapists need to use different treatment methods to train patients’ upper limbs.The treatment process is monotonous,repetitive and inefficient,and it is difficult to ensure the stability and continuity of rehabilitation training due to the personal experience and physical condition of rehabilitation therapists.In addition,in the context of an aging population,there is a huge shortage of rehabilitation therapists,and patient treatment costs will increase.The rehabilitation robot system can offer consistent rehabilitation programs,the continuity of the rehabilitation process,and make up the shortage of rehabilitation therapists,so as to better solve the above problems.A reasonable robot configuration can ensure the flexibility and movement space of rehabilitation movements,and effective rehabilitation training trajectory can ensure the adaptability and effectiveness of rehabilitation training for patients.In order to meet the rehabilitation needs of patients with upper limb motor dysfunction,a 7 degree of freedoms(Do Fs)upper limb rehabilitation robot(ULRR)with hybrid mechanism is proposed based on the human upper limb movement and stroke rehabilitation theory.The main research contents are as follows:(1)Based on the theoretical knowledge of stroke rehabilitation,combined with the anatomical and kinematic characteristics of the human upper limb,a ULRR with a hybrid mechanism was proposed.Design and study ULRR by analyzing the motion and joint activity of each joint in the human upper limb.And conduct a detailed analysis of the joint structures of the robot.By designing a gravity compensation device in the shoulder elbow mechanism,the problem of the shoulder joint overcoming heavy workload from gravity has been solved.Designed a flexible parallel hemispherical shell mechanism for the wrist based on cable-spring,which completes radial,ulnar flexion,extension,flexion and circular motion of the wrist joint by changing the length of the cable and spring.Finally,the overall structural design of the ULRR was completed.(2)Aiming at the problem that flexible parallel mechanisms containing flexible springs and hybrid mechanisms cannot be directly solved for kinematic problems,a method combining statics and kinematics was proposed to solve the kinematics of the parallel mechanism.Based on screw theory and exponential product formula,the kinematics of hybrid ULRR is analyzed and solved.The Paden-Kahan subproblem is used to solve the inverse kinematics of the hybrid mechanism.Based on Monte Carlo method,the motion space of the mechanism is solved,and the motion working space of the ULRR was obtained.The correctness of kinematics is verified by simulation analysis.Based on Lagrange method and Newton-Euler method,the dynamics models of elbow and shoulder joints and wrist joints of the robot were constructed,and the rationality of the designed robot structure was verified by simulation analysis.(3)In view of the diversity of upper limb structure of rehabilitation patients and the difficulty of adapting the movement mode of patients’ upper limb in rehabilitation training by using fixed teaching parameters,a model based on(Graph Attention Networks,GATs)introducing attention mechanism was proposed for rehabilitation track teaching learning modeling.Through learning different rehabilitation training movements demonstrated by rehabilitation doctors,the rehabilitation trajectory meeting the law of human kinematics and the law of rehabilitation medicine is generated;Based on the combination of CNNs and dynamic motion primitive method,the mapping relationship between the robot and the patient’s upper limb is established,and the robot trajectory is generated to realize the problem that different patients’ upper limb structures need to design different dynamic motion primitive parameters.(4)In order to make the rehabilitation training trajectory generated by the upper limb rehabilitation robot more suitable for patients of different rehabilitation levels,a rehabilitation evaluation model for upper limb motor function of stroke patients based on the theory of confidence rule banks was proposed.Based on the fusion of expert knowledge with the motor function state scale,an upper limb rehabilitation state evaluation model based on Belief Rule Base(BRB)was established.Research methods for updating and optimizing model parameters.On the basis of evaluating the rehabilitation status of patients,a cosine similarity method is proposed to determine the rehabilitation training trajectory.(5)Develop the experimental platform of ULRR and verify the feasibility of the designed parallel mechanism through experiments.At the same time,a control system for the ULRR was constructed,and rehabilitation movement experiments were completed.By analyzing the collected test data,the consistency between the motion performance and theoretical analysis of the ULRR and the rationality of its structure were verified.