摘要
假肢手的动作数量与轻量化之间存在矛盾关系,为兼顾两者之间的平衡,满足假肢手多动作和轻量化的要求,该研究通过分析人手的16种日常抓取动作,设计了一种合理的电机驱动结构。该结构在四指中应用了多关节同时屈曲传动,在拇指中应用了定轨迹适应性传动,掌骨使用可自动切换为弧面和平面的对称弹性串联驱动式传动,以及配置五指的自动伸展,将这些机能融合在假肢手中,仅用3个电机实现了11种假肢手动作,达到了132.1 g的轻量化设计。该文还通过肌电信号结合神经网络算法,实现了假肢手直觉控制,并验证了其具有良好的抓取稳定性和操作性。
It is still a challenge to design a hand prosthesis with a consideration of multi-motions and light weight.In this paper,by analyzing 16 commonly-used motions of human hands,a trade-off plan between weight saving and the number of motions of a hand prosthesis is studied.We determine the functions to be implemented as a constant interlock mechanism of four fingers.An adaptive mechanism is applied for the thumb,and the symmetric series elastic actuator is used for the arching of metacarpal.With such a design,a prosthetic hand was designed just using three motors embedded in the palm,which has weight of 132.1 g and could perform 11 motions.The grasping stability and operability of the hand prosthesis were confirmed with intuitive myoelectric control based on a neural network algorithm in the subject experiments.
作者
雍旭
景晓蓓
矢吹佳子
東郷俊太
横井浩史
李光林
YONG Xu;JING Xiaobei;YABUKI Yoshiko;TOGO Shunta;YOKOI Hiroshi;LI Guanglin(Shenzhen Institute of Advanced Technology,Chinese Academy of Sciences,Shenzhen 518055,China;The University of Electro-Communications,Tokyo 1828585,Japan)
出处
《集成技术》
2022年第4期31-43,共13页
Journal of Integration Technology
基金
广东省重点领域研发计划项目(2020B0909020004)
深圳市科技计划项目(CJGJZD20200617103002006)。
关键词
轻量化
多动作
适应性抓取
对称弹性串联驱动式传动
神经网络算法
仿生假肢手
light weight
multi-motion
adaptive grasping
symmetric series elastic actuator
neural network algorithm
bionic prosthetic hand