Inertial Properties in Robotic Manipulation: An Object-Level Framework

Author:

Khatib Oussama1

Affiliation:

1. Robotics Laboratory Department of Computer Science Stanford University Stanford, CA 94305

Abstract

Consideration of dynamics is critical in the analysis, design, and control of robot systems. This article presents an extensive study of the dynamic properties of several important classes of robotic structures and proposes a number of general dynamic strategies for their coordination and control. This work is a synthesis of both previous and new results developed within the task-oriented operational space formulation. Here we in troduce a unifying framework for the analysis and control of robotic systems, beginning with an analysis of inertial prop erties based on two models that independently describe the mass and inertial characteristics associated with linear and angular motions. To visualize these properties, we propose a new geometric representation, termed the belted ellipsoid, that displays the magnitudes of the mass/inertial properties directly rather than their square roots. Our study of serial macro/mini structures is based on two models of redundant mechanisms. The first is a description of the task-level dy namics that results from projecting the system dynamics into operational space. The second is a unique dynamically con sistent relationship between end-effector forces and joint torques. It divides control torques at the joint level into two dynamically decoupled vectors: torques that correspond to forces at the end effector, and torques that affect only internal motions. The analysis of inertial properties of macro-/mini- manipulator systems reveals another important characteristic: that of reduced effective inertia. We show that the effective mass/inertia of a macro-/mini-manipulator is bounded above by the mass/inertia of the mini-manipulator alone. Because mini structures have a limited range of motion, we also pro pose a dextrous dynamic coordination strategy to allow full use of the high mechanical bandwidth of the mini-structures in extended-motion operations. Finally, a study of the dy namics of parallel, multiarm structures reveals an important additive property. The effective mass and inertia of a multi arm system at some operational point are shown to be given by the sum of the effective masses and inertias associated with the object and each arm. Using this property, the mul tiarm system can be treated as a single augmented object and controlled by the total operational forces applied by the arms. Both the augmented object construct and the dynam ically consistent force/torque relationship are extended for the analysis and control of multiarm systems involving redun dancy.

Publisher

SAGE Publications

Subject

Applied Mathematics,Artificial Intelligence,Electrical and Electronic Engineering,Mechanical Engineering,Modeling and Simulation,Software

Cited by 325 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Safe physical human–robot interaction through variable impedance control based on ISO/TS 15066;International Journal on Interactive Design and Manufacturing (IJIDeM);2024-08-30

2. Unified force-impedance control;The International Journal of Robotics Research;2024-07-23

3. Impedance-based null-space control of redundant torque-controlled robot;Industrial Robot: the international journal of robotics research and application;2024-06-26

4. Optimal Design of a Mobile Spray-Painting Robot Based on Dynamic Fluctuation Performance;2024 6th International Conference on Reconfigurable Mechanisms and Robots (ReMAR);2024-06-23

5. Robot control based on motor primitives: A comparison of two approaches;The International Journal of Robotics Research;2024-06-21

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3