Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators

J. Tar, I. Rudas, J. F. Bito, O. M. Kaynak

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

Abstract

In the computed torque control of strongly coupled non-linear multivariable systems as robots the practical lack of the exact dynamic model of the system to be controlled as well as the presence of a dynamic interaction with an unmodeled environment normally means considerable difficulty. The traditional way using the Denavit-Hartenberg conventions for constructing the dynamic model and representing the unknown system parameters by itself results in considerable computational complexity. Identification of these parameters via observing the behavior of the system is far more complex task that cannot always be carried out. A novel approach of this problem based on Canonical Transformations as improvements of an initial, very simple system model was recently investigated from many aspects. Though it was able to produce certain adaptivity, its limitations also were revealed. The aim of this paper is to present a possible improvement of this method resulting in the combination of soft computing and simple numerical algorithms of uniform structure independent of the particular structure of the robot arm to be controlled. The improved method is discussed and its abilities are illustrated by simulation results.

Original languageEnglish
Title of host publicationInternational Conference on Advanced Robotics, Proceedings, ICAR
PublisherIEEE
Pages873-878
Number of pages6
Publication statusPublished - 1997
EventProceedings of the 1997 8th International Conference on Advanced Robotics, ICAR'97 - Monterey, CA, USA
Duration: Jul 7 1997Jul 9 1997

Other

OtherProceedings of the 1997 8th International Conference on Advanced Robotics, ICAR'97
CityMonterey, CA, USA
Period7/7/977/9/97

Fingerprint

Manipulators
Dynamic models
Robots
Multivariable systems
Soft computing
Torque control
Computational complexity

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Tar, J., Rudas, I., Bito, J. F., & Kaynak, O. M. (1997). Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators. In International Conference on Advanced Robotics, Proceedings, ICAR (pp. 873-878). IEEE.

Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators. / Tar, J.; Rudas, I.; Bito, J. F.; Kaynak, O. M.

International Conference on Advanced Robotics, Proceedings, ICAR. IEEE, 1997. p. 873-878.

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

Tar, J, Rudas, I, Bito, JF & Kaynak, OM 1997, Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators. in International Conference on Advanced Robotics, Proceedings, ICAR. IEEE, pp. 873-878, Proceedings of the 1997 8th International Conference on Advanced Robotics, ICAR'97, Monterey, CA, USA, 7/7/97.
Tar J, Rudas I, Bito JF, Kaynak OM. Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators. In International Conference on Advanced Robotics, Proceedings, ICAR. IEEE. 1997. p. 873-878
Tar, J. ; Rudas, I. ; Bito, J. F. ; Kaynak, O. M. / Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators. International Conference on Advanced Robotics, Proceedings, ICAR. IEEE, 1997. pp. 873-878
@inproceedings{42da4a3d1d994afead35abea7708149f,
title = "Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators",
abstract = "In the computed torque control of strongly coupled non-linear multivariable systems as robots the practical lack of the exact dynamic model of the system to be controlled as well as the presence of a dynamic interaction with an unmodeled environment normally means considerable difficulty. The traditional way using the Denavit-Hartenberg conventions for constructing the dynamic model and representing the unknown system parameters by itself results in considerable computational complexity. Identification of these parameters via observing the behavior of the system is far more complex task that cannot always be carried out. A novel approach of this problem based on Canonical Transformations as improvements of an initial, very simple system model was recently investigated from many aspects. Though it was able to produce certain adaptivity, its limitations also were revealed. The aim of this paper is to present a possible improvement of this method resulting in the combination of soft computing and simple numerical algorithms of uniform structure independent of the particular structure of the robot arm to be controlled. The improved method is discussed and its abilities are illustrated by simulation results.",
author = "J. Tar and I. Rudas and Bito, {J. F.} and Kaynak, {O. M.}",
year = "1997",
language = "English",
pages = "873--878",
booktitle = "International Conference on Advanced Robotics, Proceedings, ICAR",
publisher = "IEEE",

}

TY - GEN

T1 - Constrained canonical transformations used for representing system parameters in an adaptive control of manipulators

AU - Tar, J.

AU - Rudas, I.

AU - Bito, J. F.

AU - Kaynak, O. M.

PY - 1997

Y1 - 1997

N2 - In the computed torque control of strongly coupled non-linear multivariable systems as robots the practical lack of the exact dynamic model of the system to be controlled as well as the presence of a dynamic interaction with an unmodeled environment normally means considerable difficulty. The traditional way using the Denavit-Hartenberg conventions for constructing the dynamic model and representing the unknown system parameters by itself results in considerable computational complexity. Identification of these parameters via observing the behavior of the system is far more complex task that cannot always be carried out. A novel approach of this problem based on Canonical Transformations as improvements of an initial, very simple system model was recently investigated from many aspects. Though it was able to produce certain adaptivity, its limitations also were revealed. The aim of this paper is to present a possible improvement of this method resulting in the combination of soft computing and simple numerical algorithms of uniform structure independent of the particular structure of the robot arm to be controlled. The improved method is discussed and its abilities are illustrated by simulation results.

AB - In the computed torque control of strongly coupled non-linear multivariable systems as robots the practical lack of the exact dynamic model of the system to be controlled as well as the presence of a dynamic interaction with an unmodeled environment normally means considerable difficulty. The traditional way using the Denavit-Hartenberg conventions for constructing the dynamic model and representing the unknown system parameters by itself results in considerable computational complexity. Identification of these parameters via observing the behavior of the system is far more complex task that cannot always be carried out. A novel approach of this problem based on Canonical Transformations as improvements of an initial, very simple system model was recently investigated from many aspects. Though it was able to produce certain adaptivity, its limitations also were revealed. The aim of this paper is to present a possible improvement of this method resulting in the combination of soft computing and simple numerical algorithms of uniform structure independent of the particular structure of the robot arm to be controlled. The improved method is discussed and its abilities are illustrated by simulation results.

UR - http://www.scopus.com/inward/record.url?scp=0030679026&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030679026&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0030679026

SP - 873

EP - 878

BT - International Conference on Advanced Robotics, Proceedings, ICAR

PB - IEEE

ER -