基于高斯过程的人体步态关节力矩估计

doi:10.3969/j.issn.1006-9771.2022.08.015

《中国康复理论与实践》 ›› 2022, Vol. 28 ›› Issue (8): 989-992.doi: 10.3969/j.issn.1006-9771.2022.08.015

• 辅助技术 • 上一篇

基于高斯过程的人体步态关节力矩估计

李斯绮,张宇玲,杨建涛()

上海理工大学康复工程与技术研究所,上海市 200093

收稿日期:2022-04-19 修回日期:2022-07-07 出版日期:2022-08-25 发布日期:2022-08-30
通讯作者: 杨建涛 E-mail:yjt@usst.edu.cn
作者简介:李斯绮(2001-),女,汉族,上海市人,本科生。

Estimation of joint torques in human gait based on Gaussian process

LI Siqi,ZHANG Yuling,YANG Jiantao()

Institute of Rehabilitation Engineering and Technology, University of Shanghai for Science and Technology, Shanghai 200093, China

Received:2022-04-19 Revised:2022-07-07 Published:2022-08-25 Online:2022-08-30
Contact: YANG Jiantao E-mail:yjt@usst.edu.cn

摘要/Abstract

摘要：

目的提出一种基于高斯过程的下肢关节力矩估计方法,实现人体自然步态下下肢关节力矩精确估计。

方法根据自然步态关节力矩曲线特点,选取平方指数核函数探索关节角度和关节力矩之间的相互关系,建立基于高斯过程的数据融合模型,该模型输入为下肢关节角度,输出为关节力矩。

结果 1例健康受试者以0.8 m/s的行走速度在步态跑台行走,采用提出的高斯过程模型进行3次关节力矩预测试验,试验结果显示大部分预测的期望值都落在置信区间内,89%的预测值与实际值的r² > 0.8。

结论该方法可以较为精确地估计出关节力矩。该研究潜在应用在于优化外骨骼机器人、控制主动式假肢以及调节类人机器人的关节力矩等。

关键词: 高斯过程, 人体步态, 关节力矩

Abstract:

Objective To propose an estimation method of lower limb joints torques based on Gaussian process to achieve the accurate estimation of the lower limb joint torques in human gait.

Methods According to the characteristics of the natural gait joint torques curve, the squared exponential kernel function was selected to explore the interrelationship between the joint angle and the joint torques. A data fusion model based on Gaussian process was established. The lower limb joint angle was used as the input of the model, and the output was the joint torque.

Results One healthy subject walked on the gait running platform with a walking speed of 0.8 m/s. Three joint-torque experiments were conducted using the proposed Gaussian process. The results showed that most of the predicted values fell within the confidence intervals and 89% of the r² values were greater than 0.8.

Conclusion This method could achieve accurate estimation of joint torques. The potential application of this research is to optimize exoskeleton robots, control active prosthesis, and adjust the joint torque of humanoid robots.

Key words: Gaussian process, human gait, joint torque

中图分类号:

R496

李斯绮,张宇玲,杨建涛. 基于高斯过程的人体步态关节力矩估计[J]. 《中国康复理论与实践》, 2022, 28(8): 989-992.

LI Siqi,ZHANG Yuling,YANG Jiantao. Estimation of joint torques in human gait based on Gaussian process[J]. 《Chinese Journal of Rehabilitation Theory and Practice》, 2022, 28(8): 989-992.

图/表 3

参考文献 20

[1]	蔡飞. 基于人体步态数据的下肢外骨骼动力学建模及应用研究[D]. 郑州: 中原工学院, 2019.
	CAI F. Modeling and application of lower extremity exoskeleton dynamics based on human gait data[D]. Zhengzhou: Zhongyuan University of Technology, 2019.
[2]	刘会勇, 赵青. 下肢外骨骼助行机器人研究现状及发展趋势[J]. 机械设计与制造, 2013(8): 146-148, 151.
	LIU H Y, ZHAO Q. Research status and development trends of lower extremity exoskeleton robot walker[J]. Machine Design Manufac, 2013(8):146-148, 151.
[3]	PFEIFER S, VALLERY H, HARDEGGER M, et al. Model-based estimation of knee stiffness[J]. IEEE Trans Biomed Eng, 2012, 59(9): 2604-2612. doi: 10.1109/TBME.2012.2207895
[4]	LOSEY D P, MCDONALD C G, BATTAGLIA E, et al. A review of intent detection, arbitration, and communication aspects of shared control for physical human-robot interaction[J]. Appl Mech Rev, 2018, 70(1): 010804. doi: 10.1115/1.4039145
[5]	尹逊锋. 基于肌电融合模型的正常步态下肢关节力矩解算的研究[D]. 哈尔滨江: 哈尔滨工业大学, 2021.
	YIN X F. Research on lower limb with normal gait joint torque calculation based on EMG fusion model[D]. Harbin: Harbin Institute of Technology, 2021.
[6]	葛祎霏. 基于人体步态特性的下肢康复机器人控制系统设计[D]. 郑州: 中原工学院,2021.
	GE Y F. Control system design of lower limb rehabilitation robot based on human gait characteristics[D]. Zhengzhou: Zhongyuan University of Technology, 2021.
[7]	ZHANG J J, CHEAH C C, COLLINS S H. Experimental comparison of torque control methods on an ankle exoskeleton during human walking[C]. Seattle, WA, USA: IEEE International Conference on Robotics and Automation, 2015.
[8]	张铁, 梁骁翃. 平面关节型机器人关节力矩的卡尔曼估计[J]. 浙江大学学报(工学版), 2018, 52(5): 951-959.
	ZHANG T, LIANG X H. Kalman filter-based SCARA robot joint torque estimation[J]. J Zhejiang University (Engineering Ed), 2018, 52(5): 951-959.
[9]	EILENBERG M F, GEYER H, HERR H. Control of a powered ankle-foot prosthesis based on a neuromuscular model[J]. IEEE Trans Neural Sys Rehabil Eng, 2010, 18(2): 164-173. doi: 10.1109/TNSRE.2009.2039620
[10]	王刚. 基于人体生理信号的下肢康复机器人交互控制方法研究[D]. 长春: 长春工业大学, 2021.
	WANG G. Research on interaction control method of lower limb rehabilitation robot based on human physiological signals[D]. Changchun: Changchun University of Technology, 2021.
[11]	SARTORI M, MACULAN M, PIZZOLATO C, et al. Modeling and simulating the neuromuscular mechanisms regulating ankle and knee joint stiffness during human locomotion[J]. J Neurophysiol, 2015, 114(4): 2509-2527. doi: 10.1152/jn.00989.2014
[12]	HILL A V, PAGANINI-HILL A, ZEIGLER-HILL V. The heat of shortening and the dynamic constants of muscle[J]. Proc R Soc Lond Ser B-Biol Sci, 1938, 126(843): 136-195.
[13]	熊保平, 史武翔, 林昱, 等. 基于Hill肌肉模型的人体关节力矩智能预测[J]. 北京生物医学工程, 2021, 40(1): 11-23.
	XIONG B P, SHI W X, LIN Y, et al. Human joint moment prediction based on artificial neural network[J]. Beijing Biomed Eng, 2021, 40(1): 11-23.
[14]	秦菲菲, 赵韩, 黄康, 等. 基于UK理论的下肢关节力矩分析方法[J]. 应用力学学报, 2020, 37(5): 2141-2145, 2329-2330.
	QIN F F, ZHAO H, HUANG K, et al. Analysis of lower extremity joint torque based on UK theory[J]. J Appl Mech, 2020, 37(5): 2141-2145, 2329-2330.
[15]	VASUDEVAN S. Data fusion with Gaussian processes[J]. Robot Auton Sys, 2012, 60(12): 1528-1544. doi: 10.1016/j.robot.2012.08.006
[16]	ROBERTSON D, CALDWELL G, HAMILL J, et al. Research Methods in Biomechanics[M]. 2nd ed. Champaign: Human Kinetics, 2013.
[17]	孙立剑. 基于高斯过程的复杂光学曲面重建和多传感器数据融合方法研究[D]. 上海: 上海交通大学, 2018.
	SUN L J. Research on complex optical surface reconstruction and multi-sensor data fusion method based on Gaussian process[D]. Shanghai: Shanghai Jiao Tong University, 2018.
[18]	KOCIJAN J. Modelling and Control of Dynamic Systems Using Gaussian Process Models[M]. Cham: Springer, 2016.
[19]	BOYLE P, FREAN M. Dependent Gaussian Processes[M]. Cambridge: MIT Press, 2004: 217-224.
[20]	YANG J T, YIN Y H. Dependent-Gaussian-process-based learning of joint torques using wearable smart shoes for exoskeleton[J]. Sensors, 2020, 20(13): 3685. doi: 10.3390/s20133685

关节	第1次	第2次	第3次
踝关节	0.9129	0.9685	0.9450
膝关节	0.6278	0.8815	0.9516
髋关节	0.8668	0.9468	0.8778

基于高斯过程的人体步态关节力矩估计

Estimation of joint torques in human gait based on Gaussian process

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 3

参考文献 20

相关文章 1

Metrics

本文评价

推荐阅读 0