工作区
运动学
机制(生物学)
任务(项目管理)
适应性
欠驱动
计算机科学
反向动力学
机械手
人工智能
工程类
计算机视觉
模拟
机器人
生态学
哲学
物理
系统工程
认识论
经典力学
生物
作者
Jin Xiao-dong,Yuefa Fang,Dan Zhang,Haiqiang Zhang
出处
期刊:Journal of Mechanical Design
日期:2020-02-14
卷期号:142 (8)
被引量:8
摘要
Abstract Dexterous hands are an important end-effector of robots, but their relatively low carrying capacity, small workspace and poor task adaptability are the key factors that restrict their wide application. To overcome these shortcomings of dexterous hands, a novel Lie-group-based synthesis method that extends the 3-[P][S] parallel mechanisms (PMs) to dexterous hands is presented, and a class of three-finger dexterous hands with parallel finger structure is obtained. The multimode operation is proposed by designing a double-slider palm that provides the hands with a large workspace and high task adaptability. The operation types are presented, and the dexterous in-hand manipulations in all modes are analyzed by means of Lie group theory. In addition, the equivalent structural characteristics of pinching objects are classified to elucidate the motion types and the rotational properties of the pinched objects. The inverse kinematics of fingers is presented and is used to identify the input–output relationships. Finally, the workspaces of the fingers are determined according to the result of the inverse kinematics, and the relationships between the size and displacements of the pinched object are presented. The proposed dexterous hands overcome the problems of low carrying capability, small workspace, and weak in-hand manipulation ability that are encountered with the traditional dexterous hands, which are underactuated and are built with a series finger structure, and can be potentially applied to various application domains, such as services, industry, and rescue.
科研通智能强力驱动
Strongly Powered by AbleSci AI