Colocated and noncolocated control design via μ-synthesis for flexible manipulators

Mansour Karkoub, Gary J. Balas, Kumar Tamma

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Sensor and actuator locations play an important role in the control of flexible manipulators. Colocated and noncolocated controllers are designed using μ-synthesis technique for an experimental flexible arm testbed at the University of Minnesota. A nonlinear, dynamic model of the flexible arm is linearized to generate the control design model. Discrepancies between a linear model and the original dynamical model are modeled as uncertainties and included in the μ-framework. It is observed, theoretically and experimentally, that the addition of tip sensing dramatically improves the tip tracking performance and robustness. The robustness of the noncolocated controllers to variations in the control design model is verified theoretically and experimentally.

Original languageEnglish
Pages (from-to)3321-3325
Number of pages5
JournalProceedings of the American Control Conference
Volume5
Publication statusPublished - 1995
Externally publishedYes

Fingerprint

Flexible manipulators
Robustness (control systems)
Controllers
Testbeds
Dynamic models
Actuators
Sensors

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Colocated and noncolocated control design via μ-synthesis for flexible manipulators. / Karkoub, Mansour; Balas, Gary J.; Tamma, Kumar.

In: Proceedings of the American Control Conference, Vol. 5, 1995, p. 3321-3325.

Research output: Contribution to journalArticle

@article{497291c3482b46ef97f935a90fb070df,
title = "Colocated and noncolocated control design via μ-synthesis for flexible manipulators",
abstract = "Sensor and actuator locations play an important role in the control of flexible manipulators. Colocated and noncolocated controllers are designed using μ-synthesis technique for an experimental flexible arm testbed at the University of Minnesota. A nonlinear, dynamic model of the flexible arm is linearized to generate the control design model. Discrepancies between a linear model and the original dynamical model are modeled as uncertainties and included in the μ-framework. It is observed, theoretically and experimentally, that the addition of tip sensing dramatically improves the tip tracking performance and robustness. The robustness of the noncolocated controllers to variations in the control design model is verified theoretically and experimentally.",
author = "Mansour Karkoub and Balas, {Gary J.} and Kumar Tamma",
year = "1995",
language = "English",
volume = "5",
pages = "3321--3325",
journal = "Proceedings of the American Control Conference",
issn = "0743-1619",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Colocated and noncolocated control design via μ-synthesis for flexible manipulators

AU - Karkoub, Mansour

AU - Balas, Gary J.

AU - Tamma, Kumar

PY - 1995

Y1 - 1995

N2 - Sensor and actuator locations play an important role in the control of flexible manipulators. Colocated and noncolocated controllers are designed using μ-synthesis technique for an experimental flexible arm testbed at the University of Minnesota. A nonlinear, dynamic model of the flexible arm is linearized to generate the control design model. Discrepancies between a linear model and the original dynamical model are modeled as uncertainties and included in the μ-framework. It is observed, theoretically and experimentally, that the addition of tip sensing dramatically improves the tip tracking performance and robustness. The robustness of the noncolocated controllers to variations in the control design model is verified theoretically and experimentally.

AB - Sensor and actuator locations play an important role in the control of flexible manipulators. Colocated and noncolocated controllers are designed using μ-synthesis technique for an experimental flexible arm testbed at the University of Minnesota. A nonlinear, dynamic model of the flexible arm is linearized to generate the control design model. Discrepancies between a linear model and the original dynamical model are modeled as uncertainties and included in the μ-framework. It is observed, theoretically and experimentally, that the addition of tip sensing dramatically improves the tip tracking performance and robustness. The robustness of the noncolocated controllers to variations in the control design model is verified theoretically and experimentally.

UR - http://www.scopus.com/inward/record.url?scp=0029191221&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029191221&partnerID=8YFLogxK

M3 - Article

VL - 5

SP - 3321

EP - 3325

JO - Proceedings of the American Control Conference

JF - Proceedings of the American Control Conference

SN - 0743-1619

ER -