Research On Design And Force Control Of Massage Robot Hand For Upper Limb Rehabilitation

With the aging of China’s population,the incidence and prevalence of stroke have increased year by year.Most stroke patients suffer severe body disfunction,such as hemiplegic sensation and movement disorder,among which the share of upper limb hemiplegia has reached 50%～70%.Patients show symptoms,like upper limb flexion,muscle spasms and abnormal muscle tone,which have a negative effect on their lives and mental health.Massage is one of useful and common methods for upper limb rehabilitation.However,there is great shortage of professional massagists,besides,the research of massage robot for upper limb rehabilitation is insufficient,therefore,this topic,based on traditional massage,has designed a massage robot hand which can achieve four massage techniques for upper limb rehabilitation,so the massage robot hand has great application value.First of all,this paper analyzes four massage techniques for upper limb rehabilitation:finger pressing,finger kneading,finger pinching and finger rubbing.Based on their mechanism,the finger type and transmission mode are chosen,and the three-dimensional model of the robot hand is built.The kinematic model is established and it’s verified that the robot hand can match the size of upper limb.The relation between the cartesian space of the fingertip and the motor-driven space is obtained,based on which the motion of the finger joints is optimized by the genetic algorithm.The strength of the robot hand is verified by the static mechanical simulation analysis.Then,this paper designs the sensing system.A fingertip three-dimensional force sensor is developed.The elastic body structure and hardware circuit of the threedimensional force sensor are designed in detail,and the calibration experiment is carried out.Based on the calibration data,the least square method and BP neural network are used to decouple the dimension coupling,and the performance of the three-dimensional force sensor is verified.Besides,the thin-film pressure sensor is calibrated and the angular displacement sensor is introduced.Finally,the force control of the massage robot hand is studied.The admittance control is chosen as the force control method and its force error analysis is carried out.The influence of impedance parameters on the system is analyzed by simulating constant force tracking on a fixed position,and the inadequacy of admittance control,when environmental information changes,is verified by simulating finger pressing and rubbing.Therefore,an adaptive variable impedance method is proposed,and its stability is analyzed.By finger pressing and rubbing simulation,it’s verified that the adaptive variable impedance method is superior to the traditional admittance control.A testing platform for the massage robot hand is built,based on which finger pressing and rubbing experiments are carried out,and the results of the experiments show that the adaptive variable impedance method has excellent force tracking ability,therefore the performance of the massage robot hand is verified.