Volume 40, Number 4, August 2022
|Page(s)||764 - 770|
|Published online||30 September 2022|
Study on upper limb joint angle prediction method based on sEMG
School of Mechanical Engineering, Northwestern Polytechnical University, Xi’an 710072, China
Aiming at the problems of insufficient human-computer interaction and human-machine coupling in the rehabilitation training process, a prediction model of upper limb joint angle is proposed and verified by experiments. Firstly, a mixture vector that can well represent the motion intention of the upper limbs is obtained based on sEMG; secondly, the signal preprocessing, feature optimization and extraction of temporal eigenvalues are completed; finally, for the problems of unsatisfactory prediction accuracy and slow prediction speed of the current models in the field of motion control, the least square method (LSM) is adopted. The upper limb joint angle prediction is realized by multiplying the support vector machine (LSSVM) first. The experimental results show that the prediction model proposed in this paper can well predict the motion trajectory of the upper limb joints of the human body according to the sEMG and attitude information, effectively reduce the prediction time delay and error, and has certain advantages.
针对康复训练过程人机交互性差、人机耦合不足的问题, 提出上肢关节角度预测模型并完成实验验证。基于表面肌电信号(sEMG)获得可良好表征上肢运动意图的混合向量; 完成信号的预处理、特征优化并提取时域特征值; 针对当前运动控制领域模型预测精度不理想、预测速度较慢的问题, 采用最小二乘支持向量机(LSSVM)方法实现上肢关节角度预测。实验结果证明提出的预测模型可根据表面肌电信号与姿态信息良好地预测人体上肢关节运动轨迹, 有效减少预测时滞与误差, 在提升人机耦合性方面具有一定的优越性。
Key words: upper limb exoskeleton / rehabilitation training / sEMG / LSSVM / continuous motion estimation
关键字 : 上肢外骨骼 / 表面肌电信号 / 最小二乘支持向量机 / 连续运动估计 /
© 2022 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.