Partitioned image-based visual servo control: Some new results

Peter Corke, Seth Hutchinson, Nicholas R. Gans

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In image-based visual servo control, since control is effected with respect to the image, there is no direct control over the Cartesian velocities of the robot end effector. As a result, trajectories that the robot executes, while pleasing in the image plane, can be quite contorted in the Cartesian space. In this paper we describe a partitioned approach to visual servo control that overcomes this problem. In particular, we decouple the z-axis rotational and translational components of the control from the remaining degrees of freedom. Then, to guarantee that all features remain in the image throughout the entire trajectory, we incorporate a potential function that repels feature points from the boundary of the image plane. We illustrate the new control scheme with a variety of simulations and laboratory experiments.

Original languageEnglish (US)
Title of host publicationSensor Based Intelligent Robots - International Workshop, 2000, Selected Revised Papers
EditorsGregory D. Hager, Henrik Iskov Christensen, Horst Bunke, Rolf Klein
PublisherSpringer
Pages122-140
Number of pages19
ISBN (Print)9783540433996
DOIs
StatePublished - 2002
Event4th International Workshop on Sensor Based Intelligent Robots, 2000 - Dagstuhl Castle, Germany
Duration: Oct 15 2000Oct 20 2000

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2238
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Workshop on Sensor Based Intelligent Robots, 2000
Country/TerritoryGermany
CityDagstuhl Castle
Period10/15/0010/20/00

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Partitioned image-based visual servo control: Some new results'. Together they form a unique fingerprint.

Cite this