Value and Policy Iterations in Optimal Control and Adaptive Dynamic Programming

Dimitri P. Bertsekas

doi:10.1109/TNNLS.2015.2503980

Value and Policy Iterations in Optimal Control and Adaptive Dynamic Programming

Dimitri P. Bertsekas

Research output: Contribution to journal › Article › peer-review

143 Scopus citations

Abstract

In this paper, we consider discrete-time infinite horizon problems of optimal control to a terminal set of states. These are the problems that are often taken as the starting point for adaptive dynamic programming. Under very general assumptions, we establish the uniqueness of the solution of Bellman's equation, and we provide convergence results for value and policy iterations.

Original language	English (US)
Pages (from-to)	500-509
Number of pages	10
Journal	IEEE Transactions on Neural Networks and Learning Systems
Volume	28
Issue number	3
DOIs	https://doi.org/10.1109/TNNLS.2015.2503980
State	Published - Mar 2017
Externally published	Yes

Keywords

Dynamic programming (DP)
optimal control
policy iteration (PI)
value iteration (VI)

ASJC Scopus subject areas

Software
Computer Science Applications
Computer Networks and Communications
Artificial Intelligence

Access to Document

10.1109/TNNLS.2015.2503980

Cite this

@article{4e8f64e89ed545f08bbd56d849269bc3,

title = "Value and Policy Iterations in Optimal Control and Adaptive Dynamic Programming",

abstract = "In this paper, we consider discrete-time infinite horizon problems of optimal control to a terminal set of states. These are the problems that are often taken as the starting point for adaptive dynamic programming. Under very general assumptions, we establish the uniqueness of the solution of Bellman's equation, and we provide convergence results for value and policy iterations.",

keywords = "Dynamic programming (DP), optimal control, policy iteration (PI), value iteration (VI)",

author = "Bertsekas, {Dimitri P.}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.",

year = "2017",

month = mar,

doi = "10.1109/TNNLS.2015.2503980",

language = "English (US)",

volume = "28",

pages = "500--509",

journal = "IEEE Transactions on Neural Networks and Learning Systems",

issn = "2162-237X",

publisher = "IEEE Computational Intelligence Society",

number = "3",

}

TY - JOUR

T1 - Value and Policy Iterations in Optimal Control and Adaptive Dynamic Programming

AU - Bertsekas, Dimitri P.

PY - 2017/3

Y1 - 2017/3

N2 - In this paper, we consider discrete-time infinite horizon problems of optimal control to a terminal set of states. These are the problems that are often taken as the starting point for adaptive dynamic programming. Under very general assumptions, we establish the uniqueness of the solution of Bellman's equation, and we provide convergence results for value and policy iterations.

AB - In this paper, we consider discrete-time infinite horizon problems of optimal control to a terminal set of states. These are the problems that are often taken as the starting point for adaptive dynamic programming. Under very general assumptions, we establish the uniqueness of the solution of Bellman's equation, and we provide convergence results for value and policy iterations.

KW - Dynamic programming (DP)

KW - optimal control

KW - policy iteration (PI)

KW - value iteration (VI)

UR - http://www.scopus.com/inward/record.url?scp=85027700528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027700528&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2015.2503980

DO - 10.1109/TNNLS.2015.2503980

M3 - Article

C2 - 28055911

AN - SCOPUS:85027700528

SN - 2162-237X

VL - 28

SP - 500

EP - 509

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

IS - 3

ER -

Value and Policy Iterations in Optimal Control and Adaptive Dynamic Programming

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this