Non-randomized control of constrained Markov decision processes

被引：0

作者：

Chen, Richard C. ^{[1
]}

Feinberg, Eugene A. ^{[2
]}

机构：

[1] USN, Res Lab, Div Radar, Code 5340, Washington, DC 20375 USA

[2] SUNY Stony Brook, Dept Appl Math & Stat, Stony Brook, NY 11794 USA

来源：

2006 AMERICAN CONTROL CONFERENCE, VOLS 1-12 | 2006年 / 1-12卷

关键词：

D O I：

10.1109/ACC.2006.1656446

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper presents results concerning the optimal control of constrained Markov decision processes with expected-cost criteria using non-randomized policies. A dynamic programming approach is used to construct optimal policies. The convergence of the finite horizon value function to the infinite horizon value function is also shown. A simple example application is presented.

引用

页码：1593 / +

页数：3

共 14 条

[1]

Altman E., 1991, Annals of Operations Research, V32, P1, DOI 10.1007/BF02204825

[2]

[Anonymous], 2002, HDB MARKOV DECISION

[3] Constrained stochastic control with probabilistic criteria and search optimization [J].

Chen, RC .

2004 43RD IEEE CONFERENCE ON DECISION AND CONTROL (CDC), VOLS 1-5, 2004, :3013-3020

[4]

Chen RC, 2002, P AMER CONTR CONF, V1-6, P2014, DOI 10.1109/ACC.2002.1023931

[5] Dynamic programming equations for discounted constrained stochastic control [J].

Chen, RC ;

Blankenship, GL .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2004, 49 (05) :699-709

[6]

CHEN RC, 2005, DEF APPL SIGN PROC P

[7] Mixed risk-neutral/minimax control of discrete-time, finite-state Markov decision processes [J].

Coraluppi, SP ;

Marcus, SI .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2000, 45 (03) :528-532

[8] CONSTRAINED MARKOV DECISION-MODELS WITH WEIGHTED DISCOUNTED REWARDS [J].

FEINBERG, EA ;

SHWARTZ, A .

MATHEMATICS OF OPERATIONS RESEARCH, 1995, 20 (02) :302-320

[9] Constrained dynamic programming with two discount factors: Applications and an algorithm [J].

Feinberg, EA ;

Shwartz, A .

IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1999, 44 (03) :628-631

[10] Constrained discounted Markov decision processes and Hamiltonian Cycles [J].

Feinberg, EA .

MATHEMATICS OF OPERATIONS RESEARCH, 2000, 25 (01) :130-140

← 1 2 →