Research On The Key Technology Of Rehabilitation Robot And Assessment For Upper Limb

In our country,the degree of population aging are getting worse continuously,the number of people with upper limb paralysis is increasing year by year.However,rehabilitation physicians are relatively few,the robot assisted rehabilitation treatment is still in its infancy,also rehabilitation assessment method mainly based on the traditional time-consuming manual method.The study of home rehabilitation is imminent.Therefore,the research of rehabilitation robot technology and assessment method is of great significance.The design of rehabilitation robot,the construction of rehabilitation training virtual scene and the research of assessment method are the key problems need to be solved in the interdisciplinary field of medicine and engineering.This paper carried out research on the rehabilitation robot key technology and rehabilitation assessment method,to achieve the goal of “every disabled person has access to rehabilitation services",solve the difficulties in rehabilitation training for patients with physical movement disorder,and build a good rehabilitation training and evaluation environment.Firstly,the theoretical knowledge related to rehabilitation training and assessment is studied,the foundations of kinesiology and upper limb motor workspace are analyzed,and the related methods of rehabilitation training and assessment are explored.They lay a solid foundation for research of the technology of rehabilitation robot and assessment for upper limb.Since it takes a long time for the robot to assist the affected limb in a single rehabilitation training,the patient's posture adjustment during this period is occurred frequently,which affects the effect of rehabilitation training.Therefore,a rehabilitation training method based on human posture is proposed.A human-robot cooperative upper limb rehabilitation robot system based on 4-DOF WAM robot is constructed.According to the visual information,the patient's position changed in the horizontal direction are tracked and compensated.A hierarchical fuzzy control strategy is proposed,which mainly includes the outer fuzzy interpolation strategy and the inner fuzzy PD controller.It ensures the stability,comfort and friendliness of rehabilitation training.Experimental results show that the proposed method can successfully achieve a natural and comfortable upper limb rehabilitation training,and can improve the enthusiasm of patients to participate in the training.A 3-DOF home rehabilitation training robot is developed to solve the problems of expensive,bulky and inconvenient carrying of the rehabilitation robots at present.This robot is designed based on the principle that three axes intersect at one point,which has the advantages of simple mechanism and convenient to move.A three-dimensional force sensor is equipped at the end of the manipulator to measure the interaction forces between the affected upper limb and the robot.Virtual rehabilitation training scenes are designed to improve patients' enthusiasm for rehabilitation training.A backstepping adaptive fuzzy based impedance control is proposed for the home-based upper limb rehabilitation robot to prevent secondary injury of the affected limb.The effectiveness of the control method is verified by experiments.The size of upper limb reachable workspace reflects the upper limb motor function.The decrease of muscle tension,joint swelling and stiffness in stroke patients leads to the absence of workspace,the loss of patients' self-care ability and daily activities.A method of calculating envelope RSA of upper extremity workspace is proposed.An improved version of motion protocol is designed according to the patients' needs.Virtual interactive environment is planned.The relationship between RSA and FMA of upper limb is studied.Differences in upper limb movement angles between patients and healthy people are calculate by the DTW algorithm.Finally,the upper limb assessment model based on ANFIS is designed to evaluate the upper limb motor function.Owing to the needs of home rehabilitation training and assessment,a new multi-scene upper limb rehabilitation training and evaluation system is designed.Based on vision sensor and attitude sensor,multi-sensor fusion method is used to track the motion of patients.Multiple virtual scenes are designed to encourage patients to carry out rehabilitation training of upper limbs and trunk.A rehabilitation assessment method combining Fugl-Meyer assessment scale and upper extremity accessible workspace relative surface area is proposed,and a FMA-RSA assessment model for upper extremity motor function is established.Correlation based dynamic time warping(CBDTW)is used to solve the problem of inconsistent movement path in different patients.The experimental results show that the system is very friendly to the subjects.The rehabilitation assessment results by this system are highly correlated with the therapist.It also reveals that long-term rehabilitation training has statistical significance in improving the upper limb motor function of patients.The system has the potential to become an effective home rehabilitation training and assessment system.In order to reduce the financial pressure of patients' families,an upper limb rehabilitation system based on cloud is proposed,which connects home patients and hospital through the Internet.A simple assessment method of upper limb rehabilitation based on motion tracking is proposed.According to relevant references,five motions related to shoulder,elbow and wrist joints are selected.The smoothness and stability of the motion path of the affected limb are evaluated by velocity,acceleration logarithm and curvature logarithm.TCNN assessment model is established to evaluate the level of five selected actions.This system enables patients to carry out rehabilitation training and evaluation at home,and therapists can guide patients' rehabilitation training remotely through the network.