Design And Research Of A Lower Limb Rehabilitation Robot System Based On SEMG Signal

With the continuous intensification of population aging in China,the incidence of stroke has sharply increased,and many individuals suffer from lower limb motor dysfunction due to unhealthy lifestyles or accidents.Traditional rehabilitation methods rely on rehabilitation physicians for targeted treatment,which are inefficient and consume a large amount of manpower resources.In recent years,with the rapid development of robotics technology,the medical field has seen the emergence of rehabilitation robots aimed at assisting patients in lower limb rehabilitation training.Research in this field in China is still in the preliminary stage,with numerous critical issues that need to be addressed.In this paper,a seated and supine lower limb rehabilitation robot is designed based on surface electromyography(s EMG)signals to tackle the aforementioned problems.The main work of this paper is as follows.Firstly,considering the physiological structure of the lower limbs and the requirements of seated and supine lower limb rehabilitation training,a multi-link structure dynamic model is constructed.Based on this theoretical model,the mechanical structure of a seated and supine lower limb rehabilitation robot is designed,laying the foundation for the subsequent design of the hardware and software system of the seated and supine lower limb rehabilitation robot.Secondly,to meet the requirements of seated and supine lower limb rehabilitation training,a hardware system for the seated and supine lower limb rehabilitation robot is designed with the STM32H7A3ZI-Q microcontroller as the main control chip.This hardware system includes the main controller module,drive module,sensor module,etc.,and involves the selection and configuration of various hardware modules.Furthermore,in combination with the Brunnstrom stage treatment technique,the rehabilitation training strategies required for patients with lower limb motor dysfunction in each recovery stage are analyzed.An adaptive active/passive rehabilitation training strategy is proposed.Based on the previously established seated and supine lower limb rehabilitation robot hardware system,a control system and human-machine interaction software system for active/passive training are designed.Finally,addressing the problem of recognition of lower limb movement intentions,a neural network model based on s EMG signals is proposed as the pattern recognition algorithm for the seated and supine lower limb rehabilitation robot.Using the seated and supine lower limb rehabilitation robot software and hardware system designed earlier,three types of s EMG signals generated during seated and supine rehabilitation training are collected to establish the dataset required for model training and experimentation.Through ablative experiments and analysis on this dataset,the optimization and deployment of the algorithm model are completed.