基于肌电控制的半定制柔性康复训练手套-学报杂志社

期刊信息

合肥工业大学（自然科学版），2024年3月，第47卷第3期：289-295

DOI: 10.3969/j.issn.1003-5060.2024.03.001

作者信息

刘彩霞 $ ^{1,2} $，马菲 $ ^{2} $，余宏波 $ ^{2} $，刘贵乾 $ ^{1} $，郭旭 $ ^{2} $，黄英 $ ^{2} $

(1. 合肥工业大学物理学院，安徽合肥 230601；2. 合肥工业大学微电子学院，安徽合肥 230601)

摘要和关键词

摘要: 针对现有康复训练手套舒适性差、适用范围窄、康复训练效果不佳等问题，文章设计一种由柔性执行器和肌电控制系统组成的半定制康复训练手套。基于 Yeoh 理论建立执行器弯曲模型，借助有限元仿真对执行器气腔进行分析，研究执行器气腔压强、高度、底层厚度对弯曲角度的影响；选取柔性执行器尺寸参数，采用 3D 打印技术制作长度可调的连接部件实现半定制；利用 Myoware 传感器采集皮肤表面肌电信号同时进行滤波处理，利用 Arduino UNO 单片机执行特征提取，由支持向量机（support vector machine，SVM）算法完成运动意图识别，并依据识别结果控制气动系统驱动康复训练手套。结果表明，SVM 算法分类准确率高，所研制的半定制柔性康复训练手套结构简单，与手部贴合度高，性能可靠。该康复训练手套对主动式镜像康复训练具有积极的效果。

关键词: 肌电控制；康复训练手套；软体机器人

Authors

LIU Caixia $ ^{1,2} $, MA Fei $ ^{2} $, YU Hongbo $ ^{2} $, LIU Guiqian $ ^{1} $, GUO Xu $ ^{2} $, HUANG Ying $ ^{2} $

(1. School of Physics, Hefei University of Technology, Hefei 230601, China; 2. School of Microelectronics, Hefei University of Technology, Hefei 230601, China)

Abstract and Keywords

Abstract: Aiming at the problems of poor comfort, narrow application range and poor rehabilitation training effect of the existing rehabilitation training gloves, this paper designs semi-customized rehabilitation training gloves composed of flexible actuators and EMG control systems. Based on Yeoh theory, the actuator bending model is established, and the actuator air chamber is analyzed with the help of finite element simulation to study the influence of the actuator air chamber pressure, height and the thickness of bottom layer on the bending angle. The geometric parameters of the flexible actuator are selected, and the connecting parts with adjustable length are made by 3D printing technology to realize semi-customization. Myoware sensor is used to collect sEMG and filter it. Arduino UNO Micro Control Unit performs feature extraction, and the support vector machine (SVM) algorithm is used to recognize the movement intention of the subjects, and the pneumatic system is controlled to drive the rehabilitation training gloves according to the recognition results. The results show that the SVM algorithm has high classification accuracy. The semi-customized flexible rehabilitation training gloves have simple structure, high fit with the hand and reliable performance, which are effective in the active mirror rehabilitation training.

Keywords: EMG control; rehabilitation training gloves; soft robot

基金信息

浙江省“领雁”研发攻关计划资助项目（2022C03052）

合肥工业大学学报(自然科学版)

导航菜单

基于肌电控制的半定制柔性康复训练手套