| Hand is one of the most complex and important organs of human body,which plays an important role in people’s daily life.With the increasing number of stroke patients,the number of patients with hand motor dysfunction caused by stroke is increasing year by year.The weakness or lack of strength of the hands of patients has seriously affected the quality of life of patients.The hand rehabilitation robot can enhance the patient’s finger movement ability through periodic rehabilitation training,and it has the advantages of light weight,good flexibility,high fit with human hands,and complex and diverse rehabilitation modes,which can better assist patients in rehabilitation training.However,most of the existing hand rehabilitation robots can only carry out simple control and lack a relatively perfect force information perception and feedback mechanism.Moreover,due to the nonlinear problem of the flexible pneumatic muscles,it is difficult for the hand rehabilitation robot to achieve precise position control and force control.Therefore,this article aims to assist patients to complete more complex and refined daily behavioral tasks,and deeply researches the position and force control methods of hand rehabilitation robot driven by flexible pneumatic muscles.The main research contents are as follows:(1)Research on adaptive trajectory tracking control of hand rehabilitation robot.Due to the nonlinear and hysteresis characteristics of the flexible pneumatic muscles,it is difficult to establish an accurate mathematical model.A neural network adaptive control algorithm is designed to avoid the complex modeling process and reduce the error between the expected trajectory and the actual trajectory.On this basis,taking into account the patient’s finger jitter and uncertainty interference during the rehabilitation training process,a neural network adaptive control algorithm based on the radial basis function(RBF)observer is proposed.Adaptively control the uncertain disturbance of the system to improve the position control accuracy of the hand rehabilitation robot driven by flexural pneumatic muscles and complete the passive trajectory tracking control training.(2)Research on neural network adaptive impedance control based on force feedback.Aiming at the actual grasping requirements of force feedback hand rehabilitation robot,on the basis of position control,a neural network impedance control algorithm based on force feedback is designed to analyze the influence of impedance control parameters on the system control performance.The unknown parameters of the system are estimated by using the approximation ability of neural network to realize the force signal tracking control.Considering that the hand rehabilitation robot faces the uncertainty of the contact environment,an adaptive impedance controller is designed,and a neural network adaptive impedance control algorithm based on force feedback is proposed to improve the hand rehabilitation robot’s adaptation to the changing contact environment.(3)Robot assisted grasping control for rehabilitation.Combined with a hand rehabilitation robot driven by flexible pneumatic muscles,according to the different needs of patients,force trajectory tracking control training is designed to improve the enthusiasm of patients in rehabilitation training.Research on different rehabilitation training strategies such as passive training,active training and assisted grasping training of hand rehabilitation robot.At the same time,in order to more intuitively display the process of interaction between the hand rehabilitation robot and the patient,a grasp experiment on objects of different textures are designed,and the force required to be applied is adjusted based on the force feedback signal to complete the automatic control of objects of different materials It is suitable for grasping,and achieves a good control effect on the finger force feedback signal of the hand rehabilitation robot driven by the flexible pneumatic muscles. |
PDF Full Text Request