基于改进卷积神经网络的动力下肢假肢运动意图识别
作者:
作者单位:

1.安庆师范大学;2.中国科学院合肥智能机械研究所

作者简介:

通讯作者:

中图分类号:

TP391

基金项目:

国家自然科学基金(11475003, 61603003, 11471093); 教育部“云数融合科教创新”基金(2017A09116); 安徽省科技重大专项(18030901021); 安徽省高校领军人才团队项目; 安徽省高校优秀拔尖人才培育 资助项目(gxbjZD26) 资助.


Intent recognition of power lower-limb prosthesis based on improved convolutional neural network
Author:
Affiliation:

1.Anqing Normal University;2.Institute of Intelligent Machines, Chinese Academy of Sciences

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    传统动力下肢假肢运动意图识别算法常使用机器学习算法分类器, 在特征选择方面则需要手工提取, 随着深度学习算法在很多领域都发挥了重要作用, 将深度学习算法用在运动意图识别研究中具有重要意义. 本文算法通过在传统的卷积神经网络的基础上进行改进, 更适应于本文研究的基于短时行为样本数据的运动意图识别,同时抑制了深度学习算法应用于运动意图识别中的过拟合. 在意图识别数据集中进行滑动窗口预处理, 目的是对时间序列样本做数据增广, 扩增目标数据集能够使训练集更加丰富全面, 提高了识别的精度, 运用改进后的卷积神经网络对增广后的数据集进行特征学习与分类. 实验结果表明, 该方法在13类运动模式下的识别率达到93%.

    Abstract:

    Traditional intent recognition algorithms of power lower-limb prosthesis often use machine learning algorithm classifiers, which require manual extraction in feature selection. As deep learning algorithms play an important role in many fields, deep learning algorithms are used in motion intent recognition research It is of great significance. The algorithm in this paper is improved on the basis of the traditional convolutional neural network, and is more suitable for the motion intent recognition based on the short-term behavior sample data studied in this paper, while suppressing the application of deep learning algorithms in motion intent recognition. The sliding window preprocessing is performed on the intent recognition data set. The purpose is to augment the data of the time series samples. Amplifying the target data set can make the training set more abundant and comprehensive, improve the accuracy of recognition, and use the improved volume. The product neural network performs feature learning and classification on the augmented data set. The experimental results show that the recognition rate of the method under the 13 types of motion patterns reaches 93%

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-03-21
  • 最后修改日期:2020-09-19
  • 录用日期:2020-09-27
  • 在线发布日期:
  • 出版日期: