Radial Basis Function Neural Network-based Control Method for a Upper Limb Rehabilitation Robot

As the third disease of death and disability in the world, stroke makes a huge difficulty to patients for activities of daily living. Upper limb damage after stroke is caused by weakness, joint out of control, paralysis and abnormality. With the increasing number of patients with hemiplegia, the technology of rehabilitation robots has also developed rapidly. At the same time, the control system of the robot is continuously improved. The conventional PID control system presents many drawbacks. The biggest safety hazard through experiment testing is that the system is out of control, in order to better ensure the safety of patients. In this paper, the ANSYS finite element analysis is used to analyze the mechanical structure of the mechanical structure, and the RBF neural network control system is introduced to improve the control strategy of the upper limb rehabilitation robot, which makes up for the shortcomings and safety hazards of the conventional control system. Through the results of MATLAB simulation, the control effect of the control system is evaluated to ensure its safety and stability, so that more patients with hemiplegia can reduce the pain and get safe and effective treatment.