Analysis of PID Control and Impedance Control Based on Upper Limb Rehabilitation Training Robot

Abstract

In the process of upper limb rehabilitation training robots assisting patients in the training of the affected limbs, a stable and sensitive control system is a prerequisite to ensure efficient training results. On the basis of traditional proportional-integral-derivative (PID) control and impedance control, this paper aims to explore the current status of the application of fuzzy PID, neural network and human-computer interaction in robotic rehabilitation training, and analyze the current research with examples. Finally, this paper proposes a method that integrates human-computer interaction, fuzzy PID, neural network and deep learning. The method uses sensors to analyze the neuroelectrical signals, human-computer interaction forces, and language systems, respectively and uses FNN controllers for control and reinforcement learning. The method can provide a more stable, efficient and personalized rehabilitation training system for upper limb rehabilitation training robots. The optimization method proposed in this paper is helpful for the subsequent improvement of the upper limb rehabilitation training robot. However, the method proposed in this paper has the limitation that it is difficult to optimize the superposition of specific functions.