Computer simulation of grasping and object-manipulation by a pair of multi-degrees of freedom robot fingers
Computer simulation of grasping and object-manipulation by a pair of multi-degrees of freedom robot fingers
复制标题
一对多自由度机器人手指抓取和操纵物体的计算机模拟
DOI：
10.1109/cira.2003.1222171
复制
发表时间：
2003
期刊：
影响因子：
--
通讯作者：
S. Arimoto
中科院分区：
文献类型：
--
作者：
J. Bae;K. Tahara;Shinsuke Nakamura;S. Arimoto
本文研究了一对半球形末端刚性的多自由度机器人手指对具有两个平行平面的物体的同时抓取和定向控制的计算机仿真。首先，通过指尖与物体表面之间的四个代数约束，讨论了整个手指-物体系统中的几何关系。然后，微分代数方程表示动态的整个系统的推导。其次，提出了一种用于物体同时抓取和定向控制的感觉反馈信号，并给出了基于约束稳定化方法数值求解整个手指-物体系统闭环微分代数方程的矩阵形式的完整描述。最后，给出了对物体的抓取和定向控制的仿真结果，并讨论了控制输入信号中各参数对系统性能的影响。最后，根据肌肉生理学中已知的人体肌肉的力-速度特性，提出了增益调节的指导。
This paper is concerned with computer simulation of concurrent grasp and orientation control of an object with two parallel flat surfaces by a pair of multi-degrees of freedom robot fingers that have rigid hemispherical ends. Firstly, geometrical relationships in the overall fingers-object system are discussed through four algebraic constraints between the finger tips and surfaces of the object. Then, differential algebraic equations expressing dynamics of the overall system are derived. Secondly, a sensory feedback signal for concurrent grasp and orientation control of the object is proposed and a full description of matrix form for numerically solving the closed-loop differential algebraic equation of the overall fingers-object system on the basis of the CSM (constraint stabilization method) is presented. Thirdly, simulation results of grasping and orientation control of the object are shown and then, the effects of each term of the proposed control input signal on the system performances are discussed. Finally, a guidance of gain tuning is suggested on the basis of well-known force-velocity characteristics of human muscle in muscle physiology.