Design And Research Of Lower Limb Rehabilitation Robot Control System Based On Electromyography Signals

This study focuses on the lower limb rehabilitation robot system and aims to achieve active rehabilitation training for lower limb injury patients through their own movement intentions,starting from electromyography signals,mainly focuses on the motion intention recognition based on the support vector machine model and adaptive control strategy.The main research contents are as follows:(1)To establish a mathematical model of kinematics theory for the rehabilitation robot system,analyze the movement mechanism of the rehabilitation mechanism through geometry,verify the correctness of the model,and provide reference for subsequent control strategy research.(2)Analyze the influence factors of the dominant muscle group in the flexion and extension of the lower limb,based on the characteristics of electromyography signals.On this basis,extract the time-domain characteristic values of electromyography signals produced during rest and flexion and extension movements through sliding windows.The feasibility of the target muscle characteristic values of motion intention recognition is verified through comparative experiments.(3)Establish a support vector machine model(SVM)for motion intention recognition.On this basis,the improved grey wolf optimization algorithm(I-GWO)is used to solve the convex optimization problem of the support vector machine objective function.The feasibility of motion intention recognition is verified through experiments on the accuracy of binary classification using the SVM model.(4)For the electromyographic control system,improve a control strategy that fuses electromyographic signals with pressure signals is proposed to achieve active rehabilitation training based on motion intention.On the one hand,the control strategy uses the support vector machine model for motion pattern recognition,on the other hand,the speed of the rehabilitation mechanism is adjusted through foot pressure.Experimental results show that the designed control strategy can effectively realize active rehabilitation training.The hardware scheme of the experimental system is designed for the lower limb rehabilitation system,and software is developed to process electromyographic and pressure signals.The experimental results show that the average recognition rate of the I-GWO-SVM model can reach 92%,and the designed control strategy can effectively support motion intention recognition and electromyographic control.The research provides technical support for the development and design of lower limb rehabilitation medical systems by enterprises.