Missile defense and interceptor allocation by neuro-dynamic programming

Dimitri P. Bertsekas; Mark L. Homer; David A. Logan; Stephen D. Patek; Nils R. Sandell

doi:10.1109/3468.823480

Missile defense and interceptor allocation by neuro-dynamic programming

Dimitri P. Bertsekas, Mark L. Homer, David A. Logan, Stephen D. Patek, Nils R. Sandell

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

84 Scopus citations

Abstract

The purpose of this paper is to propose a solution methodology for a missile defense problem involving the sequential allocation of defensive resources over a series of engagements. The problem is cast as a dynamic programming/Markovian decision problem, which is computationally intractable by exact methods because of its large number of states and its complex modeling issues. We have employed a neuro-dynamic programming (NDP) framework, whereby the cost-to-go function is approximated using neural network architectures that are trained on simulated data. We report on the performance obtained using several different training methods, and we compare this performance with the optimal.

Original language	English (US)
Pages (from-to)	42-51
Number of pages	10
Journal	IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans.
Volume	30
Issue number	1
DOIs	https://doi.org/10.1109/3468.823480
State	Published - Jan 2000
Externally published	Yes

ASJC Scopus subject areas

Software
Control and Systems Engineering
Human-Computer Interaction
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/3468.823480

Cite this

@article{1e2232bfef864dee9ffe73616568c7ac,

title = "Missile defense and interceptor allocation by neuro-dynamic programming",

abstract = "The purpose of this paper is to propose a solution methodology for a missile defense problem involving the sequential allocation of defensive resources over a series of engagements. The problem is cast as a dynamic programming/Markovian decision problem, which is computationally intractable by exact methods because of its large number of states and its complex modeling issues. We have employed a neuro-dynamic programming (NDP) framework, whereby the cost-to-go function is approximated using neural network architectures that are trained on simulated data. We report on the performance obtained using several different training methods, and we compare this performance with the optimal.",

author = "Bertsekas, {Dimitri P.} and Homer, {Mark L.} and Logan, {David A.} and Patek, {Stephen D.} and Sandell, {Nils R.}",

note = "Funding Information: Manuscript received February 18, 1999; revised September 29, 1999. This work was supported by a Small Business Innovative Research (SBIR) Grant. D. P. Bertsekas is with the Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, Cambridge, MA 02139 USA (e-mail: dimitrib@mit.edu). M. L. Homer is with Biogen, Cambridge, MA 02141 USA (e-mail: homer@alum.mit.edu). D. A. Logan and N. R. Sandell are with Alphatech, Inc., Burlington, MA 01803 USA (e-mail: david.logan@alphatech.com; nils.sandell@al-phatech.com). S. D. Patek is with the Department of Systems Engineering, University of Virginia, Charlottesville, VA 22903-2442 USA (e-mail: patek@virginia.edu). Publisher Item Identifier S 1083-4427(00)00283-6.",

year = "2000",

month = jan,

doi = "10.1109/3468.823480",

language = "English (US)",

volume = "30",

pages = "42--51",

journal = "IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans",

issn = "1083-4427",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - Missile defense and interceptor allocation by neuro-dynamic programming

AU - Bertsekas, Dimitri P.

AU - Homer, Mark L.

AU - Logan, David A.

AU - Patek, Stephen D.

AU - Sandell, Nils R.

N1 - Funding Information: Manuscript received February 18, 1999; revised September 29, 1999. This work was supported by a Small Business Innovative Research (SBIR) Grant. D. P. Bertsekas is with the Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, Cambridge, MA 02139 USA (e-mail: dimitrib@mit.edu). M. L. Homer is with Biogen, Cambridge, MA 02141 USA (e-mail: homer@alum.mit.edu). D. A. Logan and N. R. Sandell are with Alphatech, Inc., Burlington, MA 01803 USA (e-mail: david.logan@alphatech.com; nils.sandell@al-phatech.com). S. D. Patek is with the Department of Systems Engineering, University of Virginia, Charlottesville, VA 22903-2442 USA (e-mail: patek@virginia.edu). Publisher Item Identifier S 1083-4427(00)00283-6.

PY - 2000/1

Y1 - 2000/1

N2 - The purpose of this paper is to propose a solution methodology for a missile defense problem involving the sequential allocation of defensive resources over a series of engagements. The problem is cast as a dynamic programming/Markovian decision problem, which is computationally intractable by exact methods because of its large number of states and its complex modeling issues. We have employed a neuro-dynamic programming (NDP) framework, whereby the cost-to-go function is approximated using neural network architectures that are trained on simulated data. We report on the performance obtained using several different training methods, and we compare this performance with the optimal.

AB - The purpose of this paper is to propose a solution methodology for a missile defense problem involving the sequential allocation of defensive resources over a series of engagements. The problem is cast as a dynamic programming/Markovian decision problem, which is computationally intractable by exact methods because of its large number of states and its complex modeling issues. We have employed a neuro-dynamic programming (NDP) framework, whereby the cost-to-go function is approximated using neural network architectures that are trained on simulated data. We report on the performance obtained using several different training methods, and we compare this performance with the optimal.

UR - http://www.scopus.com/inward/record.url?scp=0033904264&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033904264&partnerID=8YFLogxK

U2 - 10.1109/3468.823480

DO - 10.1109/3468.823480

M3 - Article

AN - SCOPUS:0033904264

SN - 1083-4427

VL - 30

SP - 42

EP - 51

JO - IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans

JF - IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans

IS - 1

ER -

Missile defense and interceptor allocation by neuro-dynamic programming

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this