Structure Design And Compliant Control Strategy Of Upper Limb Exoskeleton Rehabilitation Robot

Stroke patients with intracranial cranial nerve damage cannot establish the conduction function of the central nervous system of the brain and the motor nerve of the affected limb,and it is difficult to carry out autonomous control,which hinders the motor rehabilitation of the affected limb.Modern medical research has shown that physical motor training based on neuroplasticity can restore muscle strength,repair damaged nerve cells,and rebuild the central nervous system’s motor control of the affected limb.In recent years,exoskeleton robots have proven to be the most effective tool for treating motor dysfunction caused by neurological disorders.Based on the principles of modern rehabilitation medicine and the skeletal muscle model of upper limbs,this paper conducts in-depth research on the movement mechanism of human upper limb joints,develops a new upper limb exoskeleton rehabilitation robot suitable for wearable and multi-modal switching,and takes this as the research object,proposes an adaptive impedance training method that integrates controllable damping,and synchronously realizes the joint position of the rehabilitation robot without exceeding the limit and adaptive variable intensity interactive training.The main work contents are as follows:(1)Using the skeletal muscle model of the human upper limb to analyze the motor structure,joint activity characteristics and basic size of the upper limb,combined with the principle of ergonomics,6 motor joints in three parts of the upper limb shoulder,elbow and wrist were selected as the motor unit of the overall configuration of the exoskeleton robot.Based on this motion configuration,the forward link of the drive motor and the reverse link of the magnetorheological damper are coupled,and the dualmode drive module structure of the shoulder,elbow and wrist of the robot is designed and assembled in the 3D modeling software to complete the overall mechanical structure design and optimization of the exoskeleton.(2)The kinematics and dynamics models of the robot were analyzed,the base coordinate system and each joint coordinate system were established by using the improved D-H method,and the D-H parameter table of the connecting rod joint was determined,and the positive kinematic equation,Jacobian matrix and kinematic inverse solution of the rehabilitation robot were obtained by combining the coordinate transformation formula,and the correctness of the established kinematic model was verified by comparing the kinematic model of the 6-D0F upper limb rehabilitation robot constructed in MATLAB Robotic Toolbox and ADAMS.From the perspective of energy analysis,the dynamic model of the robot is established by using the Lagrangian function,and the numerical simulation response of each joint in ADAMS and MATLAB environment is compared and analyzed,which verifies the reliability of the dynamic model.(3)Aiming at the human-computer interaction compliance control of exoskeleton rehabilitation robot,a nested adaptive impedance control method of inner and outer rings integrated with controllable damping is proposed,and the overshoot suppression and variable intensity training of joint trajectories are realized synchronously.Firstly,through the structural mode of parallel output of magnetorheological damper and joint motor,the electromagnetic damping moment is introduced into the dynamic equation of the rehabilitation robot,and then the joint trajectory tracker is designed as the inner loop controller,which uses the joint angular velocity and angular acceleration as input to calculate the excitation current value and joint torque gain coefficient for adjusting the damping torque in real time to achieve overshoot suppression.Secondly,the adaptive trajectory generator is designed as the outer loop controller by using the Lyapunov function,which calculates and corrects the expected trajectory in real time with the interaction force,joint angular velocity and angular acceleration as inputs,and realizes adaptive and variable intensity training of humancomputer interaction force by adjusting the stiffness parameters of the adversarial model.(4)The mechanical system model and control system model of 6-DOF upper limb rehabilitation robot were established in ADAMS and Simulink,and a joint simulation platform based on Simulink and ADAMS was built,and the simulation experiment of adaptive interactive control method was verified on the basis of completing the experimental tuning of system parameters,and the feasibility and effectiveness of the proposed method were verified through the comparison and analysis of experimental results.