Please wait a minute...
工程设计学报
工程设计学报  2020, Vol. 27 Issue (5): 552-559    DOI: 10.3785/j.issn.1006-754X.2020.00.070
设计理论与方法学     
基于串联弹性原理和增广上三角分解辨识算法的踝关节阻抗估计方法
徐刚, 向馗, 唐必伟, 庞牧野
武汉理工大学 自动化学院， 湖北 武汉430070
Estimation method of ankle joint impedance based on series elastic principle and AUDI algorithm
XU Gang, XIANG Kui, TANG Bi-wei, PANG Mu-ye
School of Automation， Wuhan University of Technology， Wuhan 430070， China
 全文: PDF(1594 KB)   HTML
摘要： 研究人体关节阻抗对仿生机器人阻抗控制策略研究有重要意义。针对前期设计的踝关节阻抗测量装置（impedance detection device for ankle joint，IDDAJ）机械结构复杂及机理法建模难度大，提出了一种基于串联弹性原理和增广上三角分解辨识（augmented upper-diagonal decomposition identification，AUDI）算法的踝关节阻抗估计方法。为降低对IDDAJ驱动单元的性能要求，基于串联弹性原理，将其输入扰动中的大直流分量和动态高频分量分离；为降低建模难度，采用不受模型阶次影响的AUDI算法直接辨识IDDAJ阻抗模型，以获得更加准确的装置阻抗模型参数。根据联合阻抗测量原理，在得到IDDAJ阻抗模型后，采用AUDI算法辨识小腿肌肉不同激活程度下踝关节阻抗模型的参数。结果表明，采用AUDI算法辨识得到的IDDAJ阻抗模型的可靠性高于采用机理法得到的；在小腿肌肉放松状态下踝关节阻抗模型的参数处于同一数量级，踝关节阻抗模型的阻尼分量B、刚度分量K与小腿肌肉激活程度成正相关，这与现有文献中的结果一致。研究结果可为IDDAJ的研制和其他关节阻抗测量方法的研究提供参考。
Abstract: The research of the human joint impedance is of great significance to studying the impedance control strategy of bionic robots. In view of the complex mechanical structure of the impedance detection device for ankle joint (IDDAJ) designed in the early stage and the difficulty of modeling by the mechanism method, an estimation method of ankle joint impedance based on the series elastic principle and the augmented upper-diagonal decomposition identification (AUDI) algorithm was proposed. In order to reduce the performance requirements for the drive unit of the IDDAJ, the large DC(direct-current) component and dynamic high frequency component required in the input disturbance was separated based on the series elastic principle. In order to reduce the difficulty of modeling, the IDDAJ impedance model was directly identified by the AUDI algorithm which was not affected by the model order, so as to obtain more accurate impedance model parameters of the device. According to the principle of union impedance detection, after the impedance model of IDDAJ was obtained, the impedance model parameters of the ankle joint under different activation levels of calf muscles were identified by the AUDI algorithm. The results showed that the reliability of the impedance model of IDDAJ identified by the AUDI algorithm was higher than that obtained by the mechanism method; when the calf muscles were relaxed, the impedance model parameters of the ankle joint were in the same order of magnitude， and the damping component B and stiffness component K of the ankle joint impedance model were positively correlated with the activation level of the calf muscles, which was consistent with the results in the existing literature. The research results can provide reference for the development of IDDAJ and the research of other joint impedance detection methods.
收稿日期: 2019-12-16 出版日期: 2020-10-28
CLC:  TP 23  
基金资助: 国家自然科学基金资助项目(61603284，51575412)
通讯作者: 庞牧野(1985—)，男，山东东营人，副教授，博士，从事智能机器人研究，E-mail:pangmuye@whut.edu.cn，https://orcid.org/0000-0002-3198-1265     E-mail: pangmuye@whut.edu.cn
作者简介: 徐刚(1995—)，男，湖北随州人，硕士生，从事人体关节阻抗检测技术研究，E-mail:xgx7fen@whut.edu.cn，https://orcid.org/0000-0002-5690-3087；
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

徐刚, 向馗, 唐必伟, 庞牧野. 基于串联弹性原理和增广上三角分解辨识算法的踝关节阻抗估计方法[J]. 工程设计学报, 2020, 27(5): 552-559.

XU Gang, XIANG Kui, TANG Bi-wei, PANG Mu-ye. Estimation method of ankle joint impedance based on series elastic principle and AUDI algorithm. Chinese Journal of Engineering Design, 2020, 27(5): 552-559.

链接本文:

https://www.zjujournals.com/gcsjxb/CN/10.3785/j.issn.1006-754X.2020.00.070    或    https://www.zjujournals.com/gcsjxb/CN/Y2020/V27/I5/552

[1] 张爱云, 王吉华, 高崴, 张美娟. 基于机器视觉的VVT发动机转子缺陷检测系统设计[J]. 工程设计学报, 2021, 28(6): 776-784.
[2] 赵波, 赵海鸣, 刘晨, 胡刚. 悬立式深海钴结壳采矿头的参数化设计与优化[J]. 工程设计学报, 2021, 28(5): 559-568.
[3] 卢进南, 单德兴. 基于语义分割的火车车厢位置检测研究[J]. 工程设计学报, 2020, 27(5): 568-576.
[4] 张伟, 高慧敏. 笔管缺陷自动化检测系统设计与研究[J]. 工程设计学报, 2019, 26(3): 346-353.
[5] 李丽娟, 徐尚龙, 秦 杰. 基于图像处理技术的五金件表面缺陷检测研究[J]. 工程设计学报, 2011, 18(2): 134-138.