Cart (Loading....) | Create Account
Close category search window
 

Global minimization of the robot base reaction force during 3-D maneuvers

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Doggett, W.R. ; Syst. Integration Branch, NASA Langley Res. Center, Hampton, VA, USA ; Messner, W.C. ; Jer-Nan Juang

Provides closed-form equations parameterizing the C2 smooth path that globally minimizes the Euclidean norm of a robot's peak base reaction force while avoiding obstacles during 3D maneuvers in a gravity-free environment. Also, describes a computationally efficient technique that leads to a path typically having a peak force within 5% of the optimal path. The equations used to define the robot's motion are formulated after mapping the initial configuration, final (or goal) Cartesian location, and obstacles into a new space, the center of mass (CM) space. This is a Cartesian-like space that allows direct application of many existing control techniques, such as resolved rate control. In the CM space, a series of path segments guide the robot around the obstacles. Solving a system of equations based on these segments for boundary condition dependent constants determines the path. Currently, closed-form equations are unavailable for the boundary dependent constants, preventing exact determination of the globally optimal path. This paper introduces a procedure for locating the optimal path. Its final step uses sequential quadratic programming to locate boundary dependent constants. The equation formulations assume that the initial configuration of the robot is known and that the robot mass and obstacle positions are constant during the maneuver. The method developed has direct applicability to redundant and nonredundant robots. A detailed example, based on a nonredundant robot avoiding a single obstacle, illustrates the concepts presented

Published in:

Robotics and Automation, IEEE Transactions on  (Volume:16 ,  Issue: 6 )

Date of Publication:

Dec 2000

Need Help?


IEEE Advancing Technology for Humanity About IEEE Xplore | Contact | Help | Terms of Use | Nondiscrimination Policy | Site Map | Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest professional association for the advancement of technology.
© Copyright 2014 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.