Design of the Passive Joints of Underactuated Modular Soft Hands for Fingertip Trajectory Tracking

Design of the Passive Joints of Underactuated Modular Soft Hands for Fingertip Trajectory Tracking
复制标题

DOI:
10.1109/lra.2017.2718099
复制
发表时间:
2017-10-01
影响因子:
5.2
通讯作者:
Prattichizzo, Domenico
Prattichizzo, Domenico
中科院分区:
计算机科学2区
文献类型:
--
作者:
Salvietti, Gionata;Hussain, Irfan;Prattichizzo, Domenico

文献摘要

被引文献

相似文献

在这封信中,我们提出了一种设计肌腱驱动欠驱动手的方法,其指尖在驱动时可以跟踪预定义的轨迹。我们专注于由可变形关节和刚性连杆组成的被动柔顺手。我们首先介绍一个确定合适的关节刚度和肌腱布线的程序,然后展示机器人欠驱动手指的可能实现。肌腱驱动的机器人手指的运动学和静动学分析对于定义手指关节的整体刚度值是必要的。对构成每个被动接头的元件进行结构分析,可以定义刚度与接头的主要尺寸和材料特性之间的关系。我们以机器人 Soft-SixthFinger 作为案例研究,通过仿真和实验验证了所提出的框架。 Soft-SixthFinger 是一款可穿戴机器人,用于为手部麻痹的患者进行抓取补偿。我们证明,当关节刚度和肌腱走线设计得当时,可以实现不同的指尖轨迹。此外,我们还证明,当设计特定的手指弯曲轨迹时,该设备能够抓取更广泛的物体。所提出的框架是通用的,可以应用于具有任意数量的手指和每个手指的关节的机器人手。模块化方法还允许用户根据特定任务或轨迹轻松定制手。
In this letter, we propose a method to design tendon-driven underactuated hands whose fingertips can track a predefined trajectory, when actuated. We focus on passively compliant hands composed of deformable joints and rigid links. We first introduce a procedure to determine suitable joints stiffness and tendon routing, then a possible realization of a robotic underactuated finger is shown. The kinematic and kinetostatic analysis of a tendon-driven robotic finger is necessary to define the overall stiffness values of the finger joints. A structural analysis of the element constituting each passive joint allowed to define a relation between the stiffness and joint's main dimensional and material properties. We validated the proposed framework both in simulation and with experiments using the robotic Soft-SixthFinger as a case study. The Soft-SixthFinger is a wearable robot for grasping compensation in patients with a paretic hand. We demonstrated that different fingertip trajectories can be achieved when joint stiffness and tendon routing are properly designed. Moreover, we demonstrated that the device is able to grasp a wider set of objects when a specific finger flexion trajectory is designed. The proposed framework is general and can be applied to robotic hands with an arbitrary number of fingers and joints per finger. The modular approach furthermore allows the user to easily customize the hand according to specific tasks or trajectories.