摘要
随着国防装备发展及人口老龄化加剧,战场士兵、部队后勤及老年人口等对下肢柔性外骨骼的需求急剧升高,进而研制了穿戴性好、人机相容性高、助力性能优的便携式踝助力外衣系统。基于踝关节生理特征与人体下肢参数,对踝助力外衣本体和驱动单元进行工效学设计和实物研制,并将驱动单元放置于腰部,尽可能降低小腿和足部的附加质量。基于踝关节生物力矩特征,建立期望辅助力模型,采用实验测量的办法,获得助力外衣及局部软组织的系统刚度,将期望辅助力模型转换为驱动单元的位移模型。基于人体下肢步态的周期性,设计以迭代学习控制算法为核心的控制器,通过迭代学习历史时刻的输出误差,对期望辅助力模型进行不断追踪。为评估助力外衣的助力效果,通过测量穿戴者的跖屈肌肉群肌电信号,分析肌肉激活程度和疲劳变化:在踝助力外衣的辅助下,不同行走速度的肌肉激活程度均明显下降,以腓肠肌外侧变化最为显著,腓肠肌内侧肌肉激活程度平均下降了9.62%±1.12%;平均功率频率和中位频率的变化斜率绝对值均得到一定程度的下降,肌肉疲劳速率下降、疲劳程度降低,进而下肢行走耐力增强。
With the development of national defense equipment and the aggravation of population aging,the demand of lower limb flexible exoskeletons for battlefield soldiers,army logistics and the elderly population has risen sharply.Based on this urgent demand,a portable ankle assistance Exosuit system with good wearability,high human-machine compatibility and excellent assistance performance has been developed.According to the physiological characteristics of the ankle joint and the parameters of lower limbs,the ergonomic design and development of the Exosuit body and driving unit are achieved where the driving unit is placed at the waist to reduce the added mass of legs and feet as much as possible.Based on the biological torque of the ankle joint,the model of expected auxiliary force is established.The system stiffness of the assistance Exosuit and local soft tissue is obtained by the experimental measurement.Then the established model is transformed into the displacement model of the driving unit.Based on the gait periodicity of lower limbs,a controller with iterative learning control algorithm as the core is designed to track the force model through iteratively learning the output errors of historical time.In order to evaluate the assistance performance of the Exosuit system,the degree of muscle activation and fatigue changes were analyzed by measuring the EMG signals of the plantarflexion muscles.Under the condition of Power-ON,the degree of muscle activation under three walking speeds decreased significantly,especially lateral gastrocnemius muscle.The muscle activation degree of the medial gastrocnemius muscle decreased by an average of 9.62%±1.12%.The absolute value of the change slope of mean power frequency(MPF)and median frequency(MF)descended obviously.Hence,the rate and degree of muscle fatigue decreased and the walking endurance of lower limbs increased.
作者
张雷雨
叶土仙
高翔
张斐然
李剑锋
苏鹏
ZHANG Leiyu;YE Tuxian;GAO Xiang;ZHANG Feiran;LI Jianfeng;SU Peng(Beijing Key Laboratory of Advanced Manufacturing Technology,Beijing University of Technology,Beijing 100124;Wuhan Second Ship Design and Research Institute,Wuhan 430064;School of Electromechanical Engineering,Beijing Information Science and Technology University,Beijing 100192)
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2023年第17期67-78,共12页
Journal of Mechanical Engineering
基金
北京市自然科学基金(3202003,19L2018)
国家重点研发计(2018YFB1307004)
国家自然科学基金(52005045)资助项目
关键词
柔性外骨骼
踝关节助力
迭代学习控制
肌电信号:性能评估
soft exoskeleton
ankle assistance
iterative learning control
EMG signal
performance evaluation