A Lorentz-group based adaptive control for electro-mechanical systems

József K. Tar, Imre J. Rudas, Karel Jezernik, János F. Bitó, Seppo J. Torvinen

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

1 Citation (Scopus)

Abstract

An efficient approach invented for the adaptive control of approximately and partially known mechanical systems recently was presented. Like traditional soft computing it uses "uniform structures" for modeling, but these structures are obtained from certain Lie groups as the Symplectic Group or the Generalized Lorentz Group. This approach considerably reduces the number of free parameters in the model in comparison e.g. with a neural network based model and replaces the obscure process of parameter tuning or "learning" with simple, lucid, and explicit algebraic operations of limited steps. Till now its efficiency was investigated for mechanical uncertainties and external dynamic interactions. In the present paper the behavior of the electric drives also are included in the investigations: a one degree of freedom mechanical system, a pendulum driven by a DC motor in a computed torque control is investigated via simulation. The necessary torque is calculated from a formal primitive mechanical model and an adaptivity rule. For adaptivity real number scaling and the Generalized Lorentz Group's elements are used. It is concluded that a single adaptive loop can compensate for the mechanical and the electrical uncertainties simultaneously.

Original languageEnglish
Title of host publicationIEEE International Conference on Intelligent Robots and Systems
Pages2045-2050
Number of pages6
Volume4
Publication statusPublished - 2001
Event2001 IEEE/RSJ International Conference on Intelligent Robots and Systems - Maui, HI, United States
Duration: Oct 29 2001Nov 3 2001

Other

Other2001 IEEE/RSJ International Conference on Intelligent Robots and Systems
CountryUnited States
CityMaui, HI
Period10/29/0111/3/01

Fingerprint

Lie groups
Soft computing
Electric drives
DC motors
Torque control
Pendulums
Torque
Tuning
Neural networks
Uncertainty

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Tar, J. K., Rudas, I. J., Jezernik, K., Bitó, J. F., & Torvinen, S. J. (2001). A Lorentz-group based adaptive control for electro-mechanical systems. In IEEE International Conference on Intelligent Robots and Systems (Vol. 4, pp. 2045-2050)

A Lorentz-group based adaptive control for electro-mechanical systems. / Tar, József K.; Rudas, Imre J.; Jezernik, Karel; Bitó, János F.; Torvinen, Seppo J.

IEEE International Conference on Intelligent Robots and Systems. Vol. 4 2001. p. 2045-2050.

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

Tar, JK, Rudas, IJ, Jezernik, K, Bitó, JF & Torvinen, SJ 2001, A Lorentz-group based adaptive control for electro-mechanical systems. in IEEE International Conference on Intelligent Robots and Systems. vol. 4, pp. 2045-2050, 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems, Maui, HI, United States, 10/29/01.
Tar JK, Rudas IJ, Jezernik K, Bitó JF, Torvinen SJ. A Lorentz-group based adaptive control for electro-mechanical systems. In IEEE International Conference on Intelligent Robots and Systems. Vol. 4. 2001. p. 2045-2050
Tar, József K. ; Rudas, Imre J. ; Jezernik, Karel ; Bitó, János F. ; Torvinen, Seppo J. / A Lorentz-group based adaptive control for electro-mechanical systems. IEEE International Conference on Intelligent Robots and Systems. Vol. 4 2001. pp. 2045-2050
@inproceedings{ae3ae50fea4f474f8a71951508fabb47,
title = "A Lorentz-group based adaptive control for electro-mechanical systems",
abstract = "An efficient approach invented for the adaptive control of approximately and partially known mechanical systems recently was presented. Like traditional soft computing it uses {"}uniform structures{"} for modeling, but these structures are obtained from certain Lie groups as the Symplectic Group or the Generalized Lorentz Group. This approach considerably reduces the number of free parameters in the model in comparison e.g. with a neural network based model and replaces the obscure process of parameter tuning or {"}learning{"} with simple, lucid, and explicit algebraic operations of limited steps. Till now its efficiency was investigated for mechanical uncertainties and external dynamic interactions. In the present paper the behavior of the electric drives also are included in the investigations: a one degree of freedom mechanical system, a pendulum driven by a DC motor in a computed torque control is investigated via simulation. The necessary torque is calculated from a formal primitive mechanical model and an adaptivity rule. For adaptivity real number scaling and the Generalized Lorentz Group's elements are used. It is concluded that a single adaptive loop can compensate for the mechanical and the electrical uncertainties simultaneously.",
author = "Tar, {J{\'o}zsef K.} and Rudas, {Imre J.} and Karel Jezernik and Bit{\'o}, {J{\'a}nos F.} and Torvinen, {Seppo J.}",
year = "2001",
language = "English",
volume = "4",
pages = "2045--2050",
booktitle = "IEEE International Conference on Intelligent Robots and Systems",

}

TY - GEN

T1 - A Lorentz-group based adaptive control for electro-mechanical systems

AU - Tar, József K.

AU - Rudas, Imre J.

AU - Jezernik, Karel

AU - Bitó, János F.

AU - Torvinen, Seppo J.

PY - 2001

Y1 - 2001

N2 - An efficient approach invented for the adaptive control of approximately and partially known mechanical systems recently was presented. Like traditional soft computing it uses "uniform structures" for modeling, but these structures are obtained from certain Lie groups as the Symplectic Group or the Generalized Lorentz Group. This approach considerably reduces the number of free parameters in the model in comparison e.g. with a neural network based model and replaces the obscure process of parameter tuning or "learning" with simple, lucid, and explicit algebraic operations of limited steps. Till now its efficiency was investigated for mechanical uncertainties and external dynamic interactions. In the present paper the behavior of the electric drives also are included in the investigations: a one degree of freedom mechanical system, a pendulum driven by a DC motor in a computed torque control is investigated via simulation. The necessary torque is calculated from a formal primitive mechanical model and an adaptivity rule. For adaptivity real number scaling and the Generalized Lorentz Group's elements are used. It is concluded that a single adaptive loop can compensate for the mechanical and the electrical uncertainties simultaneously.

AB - An efficient approach invented for the adaptive control of approximately and partially known mechanical systems recently was presented. Like traditional soft computing it uses "uniform structures" for modeling, but these structures are obtained from certain Lie groups as the Symplectic Group or the Generalized Lorentz Group. This approach considerably reduces the number of free parameters in the model in comparison e.g. with a neural network based model and replaces the obscure process of parameter tuning or "learning" with simple, lucid, and explicit algebraic operations of limited steps. Till now its efficiency was investigated for mechanical uncertainties and external dynamic interactions. In the present paper the behavior of the electric drives also are included in the investigations: a one degree of freedom mechanical system, a pendulum driven by a DC motor in a computed torque control is investigated via simulation. The necessary torque is calculated from a formal primitive mechanical model and an adaptivity rule. For adaptivity real number scaling and the Generalized Lorentz Group's elements are used. It is concluded that a single adaptive loop can compensate for the mechanical and the electrical uncertainties simultaneously.

UR - http://www.scopus.com/inward/record.url?scp=0035560166&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035560166&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0035560166

VL - 4

SP - 2045

EP - 2050

BT - IEEE International Conference on Intelligent Robots and Systems

ER -