Aposteriori polynomial fitting method for the general optimal control problem

Y. İstefanopulos; M. A. Eyler

doi:10.1109/IMC.1990.687417

Aposteriori polynomial fitting method for the general optimal control problem

Eksin, Y. İstefanopulos, M. A. Eyler

Department of Control and Automation Engineering

Research output: Contribution to conference › Paper › peer-review

Abstract

When the general optimal control problem is considered for arbitrary initial conditions, one finally arrives to the well-known Hamilton-Jaeobi-Bellman (HJB) equation, which is quite difficult to solve. In fact, it is rarely feasible to solve HJB equation for a nonlienar system of any practical siginificance; and hence, the development of exact explicit optimal feedback control schemes is usually unattainable for nonlienar systems. Here, a new method is presented which relies on the performance index sensitivity with respect to the changes in the controller parameter vector. This method gives rise to a sub-optimal control strategy even for nonlienar system dynamics and nonquadratic performance index.

Original language	English
Pages	761-764
Number of pages	4
DOIs	https://doi.org/10.1109/IMC.1990.687417
Publication status	Published - 1990
Event	1990 IEEE International Workshop on Intelligent Motion Control, IMC 1990 - Istanbul, Turkey Duration: 20 Aug 1990 → 22 Aug 1990

Conference

Conference	1990 IEEE International Workshop on Intelligent Motion Control, IMC 1990
Country/Territory	Turkey
City	Istanbul
Period	20/08/90 → 22/08/90

Bibliographical note

Publisher Copyright:
© 1991 IEEE.

Keywords

Nonlinear control systems
Numerical analysis
Optimization
Performance index sensitivity
Suboptimal control

Access to Document

10.1109/IMC.1990.687417

Cite this

@conference{37541ddd4e24433ebd5944610d998fc0,

title = "Aposteriori polynomial fitting method for the general optimal control problem",

abstract = "When the general optimal control problem is considered for arbitrary initial conditions, one finally arrives to the well-known Hamilton-Jaeobi-Bellman (HJB) equation, which is quite difficult to solve. In fact, it is rarely feasible to solve HJB equation for a nonlienar system of any practical siginificance; and hence, the development of exact explicit optimal feedback control schemes is usually unattainable for nonlienar systems. Here, a new method is presented which relies on the performance index sensitivity with respect to the changes in the controller parameter vector. This method gives rise to a sub-optimal control strategy even for nonlienar system dynamics and nonquadratic performance index.",

keywords = "Nonlinear control systems, Numerical analysis, Optimization, Performance index sensitivity, Suboptimal control",

author = "Eksin and Y. İstefanopulos and Eyler, {M. A.}",

note = "Publisher Copyright: {\textcopyright} 1991 IEEE.; 1990 IEEE International Workshop on Intelligent Motion Control, IMC 1990 ; Conference date: 20-08-1990 Through 22-08-1990",

year = "1990",

doi = "10.1109/IMC.1990.687417",

language = "English",

pages = "761--764",

}

TY - CONF

T1 - Aposteriori polynomial fitting method for the general optimal control problem

AU - Eksin,

AU - İstefanopulos, Y.

AU - Eyler, M. A.

PY - 1990

Y1 - 1990

N2 - When the general optimal control problem is considered for arbitrary initial conditions, one finally arrives to the well-known Hamilton-Jaeobi-Bellman (HJB) equation, which is quite difficult to solve. In fact, it is rarely feasible to solve HJB equation for a nonlienar system of any practical siginificance; and hence, the development of exact explicit optimal feedback control schemes is usually unattainable for nonlienar systems. Here, a new method is presented which relies on the performance index sensitivity with respect to the changes in the controller parameter vector. This method gives rise to a sub-optimal control strategy even for nonlienar system dynamics and nonquadratic performance index.

AB - When the general optimal control problem is considered for arbitrary initial conditions, one finally arrives to the well-known Hamilton-Jaeobi-Bellman (HJB) equation, which is quite difficult to solve. In fact, it is rarely feasible to solve HJB equation for a nonlienar system of any practical siginificance; and hence, the development of exact explicit optimal feedback control schemes is usually unattainable for nonlienar systems. Here, a new method is presented which relies on the performance index sensitivity with respect to the changes in the controller parameter vector. This method gives rise to a sub-optimal control strategy even for nonlienar system dynamics and nonquadratic performance index.

KW - Nonlinear control systems

KW - Numerical analysis

KW - Optimization

KW - Performance index sensitivity

KW - Suboptimal control

UR - http://www.scopus.com/inward/record.url?scp=85068164226&partnerID=8YFLogxK

U2 - 10.1109/IMC.1990.687417

DO - 10.1109/IMC.1990.687417

M3 - Paper

AN - SCOPUS:85068164226

SP - 761

EP - 764

T2 - 1990 IEEE International Workshop on Intelligent Motion Control, IMC 1990

Y2 - 20 August 1990 through 22 August 1990

ER -

Aposteriori polynomial fitting method for the general optimal control problem

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this