Research On Control Strategy Of Upper Limb Motion System Of Rehabilitation Robot

With the development of modern rehabilitation medicine,it has been an attractive research focus on applying the robotics-aided training system in nerve centers rehabil-itation.This dissertation based on the rehabilitation medicine principle and advanced control theory,and combining the physician experience with ergonomic design,combin-ing information collection method with man-machine interaction technique,combing the off-line programming and on-line intelligence learning,in order to solve the following problems:the optimal design of rehabilitation training trajectory,the estimation of toque parameters,the intelligence learning in passive/active training,the safety and effective-ness in active training.Finally,a complete set of robot upper limb motion system control strategy is evolved.The main research contents are outlined as follows:Characteristics and model establishment of upper limb spasticity.In rehabili-tation training,the most security risk-spasticity,was studied based on rehabilitation medicine.Based on human skeletal muscle system model,the model of upper limb spas-ticity in stroke patients with hemiplegia is established.And point out that the upper limb spasticity torque is a random variables with uncertainty.Combining the theory of random estimation with rehabilitation medicine,the estimation of random parameters in the model of spasticity torque is given by empirical Bayes estimation,and its convergence is proved.Real-time estimation of human-robot interaction torque.Considering that the existing impedance toque modeling methods can only provide part information of the end joint,a high order dynamic impedance model is used to describe the human-robot inter-action torque.Then,a nonlinear human-robot interaction model for active training stage is established The interaction torque estimation algorithm based on integral transform is proposed,and the convergence condition of the estimation algorithm is given,which can guarantees an on-line quick(exponentially convergent)estimation according to a speci-fied accuracy.Intelligent optimization algorithm design of upper limb rehabilitation trajec-tory.Based on summarizing the rehabilitation medicine and ergonomic principle,a se-ries of systemic indexes are proposed by integrating ergonomic index,patient pain in-dex,smoothness index Minimum work criterion index.Then the problem of rehabilita-tion training trajectory programming is transformed into a problem of nonlinear multi-objective optimization.The improved artificial immune algorithm is used to solve the multi-objective optimization problem of the trajectory of rehabilitation of the upper limb.It can guarantee the diversity of the population and effectively prevent premature conver-gence of the population.The algorithm can quickly search for the global optimal solution,and gives the training trajectory which meets the requirements of rehabilitation quantifi-cation and the motion characteristics of human upper limb.Nonparametric uncertain adaptive iterative learning control strategy in passive training stage.In the passive training stage,the unknown disturbance and the muscular tension uncertainty in the human-robot interaction system are considered.A nonpara-metric uncertain adaptive iterative learning control strategy is proposed to ensure that the robot moves along a predetermined trajectory smoothly.Using the correction strategy of reference trajectory and iterative error,the variable reference trajectory and the initial error uncertainty in the upper limb rehabilitation robot system are solved.The adaptive iterative learning mechanism can update the parameters of the controller in real time.It ensures the global stability of human-robot interaction system and the reliable tracking of the trajectory of rehabilitation training.Simulation experiments verify the effectiveness of the proposed control strategy.Adaptive double iteration optimal control strategy in active training stage.In view of the active torque and the spasticity torque in the active training stage,this pa-per proposes a double iterative optimal control strategy,which can guarantee the safe accomplishment and optimal modification of the training treatment.The control strategy employs a double iterative control structure,learning controller and optimal controller are iterating at the same time.Not only can deal with potential actuator failures,avoiding the second damage,but also can provide the appropriate help to optimize the limb active torque and spasticity torque.This control strategy can balance the effectiveness and safety in active rehabilitation training to ensure the rehabilitation effect and safety.Simulation experiments verify the effectiveness of the proposed control strategy.