中国组织工程研究

中国组织工程研究 ›› 2012, Vol. 16 ›› Issue (30): 5520-5523.doi: 10.3969/j.issn.2095-4344.2012.30.003

• 数字化骨科 digital orthopedics • 上一篇    下一篇

三维步态分析检测基于BP神经网络的膝踝运动

瞿晓娜1，2，张腾宇2，王喜太2   

  1. 1河北工业大学，天津市 300130；
    2国家康复辅具研究中心，北京市 100176
  • 收稿日期:2011-12-11 修回日期:2012-02-07 出版日期:2012-07-22 发布日期:2012-07-22
  • 作者简介:瞿晓娜★，女，1986年生，山东省潍坊市人，汉族，2012年河北工业大学毕业，硕士，主要从事计算机控制与工程应用的研究。 qxn_1987@126.com

Three-dimensional gait analysis of the athletic relationship between knee and ankle based on back-propagation neural network

Qu Xiao-na1, 2, Zhang Teng-yu2, Wang Xi-tai2   

  1. 1Hebei University of Technology, Tianjin 300130, China;
    2National Research Center for Rehabilitation Technical Aids, Beijing 100176, China
  • Received:2011-12-11 Revised:2012-02-07 Online:2012-07-22 Published:2012-07-22
  • About author:Qu Xiao-na★, Master, Hebei University of Technology, Tianjin 300130, China; National Research Center for Rehabilitation Technical Aids, Beijing 100176, China qxn_1987@126.com

PDF

422

被引次数

1

摘要：

背景：虽然在人体步态方面已有大量研究，但针对膝踝协调运动的研究很少。
目的：用BP神经网络分析膝踝协调运动关系。
方法：利用三维步态分析系统检测了30名健康志愿者以快、中、慢3种步速行走时的步态数据，进行统计分析；并通过建立BP神经网络预测数据，同时对膝踝协调的控制方法进行探讨。
结果与结论：不同的人步态不同，但BP神经网络预测所得曲线与实验基本一致，证实了用BP神经网络做膝踝运动关系预测的合理性和可行性，很好的研究了膝踝协调性，给全智能膝踝协调控制假肢的研发提供了理论依据。

关键词: 膝关节, 踝关节, 步态, BP神经网络, 假肢

Abstract:

BACKGROUND: The study of the relationship between knee and ankle is poor although there have been a lot of researches on gait of people.
OBJECTIVE: To analyze the relationship between knee and ankle based on back-propagation network.
METHODS: The gaits of 30 healthy young people walking at fast, normal, and low speeds separately were detected by three-dimensional gait analysis system, and the gait data were investigated and analyzed. The data were predicted through the establishment of the back-propagation neural network and the knee-ankle coordination control method was explored.
RESULTS AND CONCLUSION: Different people had different gaits, but the curve obtained by back-propagation neural network was similar with the experimental curve. The paper investigated and validated that it was viable and reasonable to forecast the kinematic relationship between the knee and ankle by back-propagation network system. The paper studied the coordination between the knee and ankle well, and provided the theory foundation for the design of the intelligent prostheses.

中图分类号: