BN+BN: Behavior network with bayesian network for intelligent agent

Kyung Joong Kim; Sung Bae Cho

doi:10.1007/978-3-540-24581-0_84

BN+BN: Behavior network with bayesian network for intelligent agent

Kyung Joong Kim, Sung Bae Cho

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

In the philosophy of behavior-based robotics, design of complex behavior needs the interaction of basic behaviors that are easily implemented. Action selection mechanism selects the most appropriate behavior among them to achieve goals of robot. Usually, robot might have one or more goals that conflict each other and needs a mechanism to coordinate them. Bayesian network represents the dependencies among variables with directed acyclic graph and infers posterior probability using prior knowledge. This paper proposes a method to improve behavior network, action selection mechanism that uses the graph of behaviors, goals and sensors with activation spreading, using goal inference mechanism of Bayesian network learned automatically. Experimental results on Khepera mobile robot show that the proposed method can generate more appropriate behaviors.

Original language	English
Title of host publication	AI 2003
Subtitle of host publication	Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings
Editors	Tamas D. Gedeon, Lance Chun Che Fung, Tamas D. Gedeon
Publisher	Springer Verlag
Pages	979-991
Number of pages	13
ISBN (Print)	9783540206460
DOIs	https://doi.org/10.1007/978-3-540-24581-0_84
Publication status	Published - 2003
Event	16th Australian Conference on Artificial Intelligence, AI 2003 - Perth, Australia Duration: 2003 Dec 3 → 2003 Dec 5

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	2903
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	16th Australian Conference on Artificial Intelligence, AI 2003
Country/Territory	Australia
City	Perth
Period	03/12/3 → 03/12/5

Bibliographical note

Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 2003.

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-540-24581-0_84

Cite this

Kim, K. J., & Cho, S. B. (2003). BN+BN: Behavior network with bayesian network for intelligent agent. In T. D. Gedeon, L. C. C. Fung, & T. D. Gedeon (Eds.), AI 2003: Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings (pp. 979-991). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2903). Springer Verlag. https://doi.org/10.1007/978-3-540-24581-0_84

Kim, Kyung Joong ; Cho, Sung Bae. / BN+BN : Behavior network with bayesian network for intelligent agent. AI 2003: Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings. editor / Tamas D. Gedeon ; Lance Chun Che Fung ; Tamas D. Gedeon. Springer Verlag, 2003. pp. 979-991 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{c1dd23c9c5e44a109e4a262f497c8517,

title = "BN+BN: Behavior network with bayesian network for intelligent agent",

abstract = "In the philosophy of behavior-based robotics, design of complex behavior needs the interaction of basic behaviors that are easily implemented. Action selection mechanism selects the most appropriate behavior among them to achieve goals of robot. Usually, robot might have one or more goals that conflict each other and needs a mechanism to coordinate them. Bayesian network represents the dependencies among variables with directed acyclic graph and infers posterior probability using prior knowledge. This paper proposes a method to improve behavior network, action selection mechanism that uses the graph of behaviors, goals and sensors with activation spreading, using goal inference mechanism of Bayesian network learned automatically. Experimental results on Khepera mobile robot show that the proposed method can generate more appropriate behaviors.",

author = "Kim, {Kyung Joong} and Cho, {Sung Bae}",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2003.; 16th Australian Conference on Artificial Intelligence, AI 2003 ; Conference date: 03-12-2003 Through 05-12-2003",

year = "2003",

doi = "10.1007/978-3-540-24581-0_84",

language = "English",

isbn = "9783540206460",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "979--991",

editor = "Gedeon, {Tamas D.} and Fung, {Lance Chun Che} and Gedeon, {Tamas D.}",

booktitle = "AI 2003",

address = "Germany",

}

Kim, KJ & Cho, SB 2003, BN+BN: Behavior network with bayesian network for intelligent agent. in TD Gedeon, LCC Fung & TD Gedeon (eds), AI 2003: Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2903, Springer Verlag, pp. 979-991, 16th Australian Conference on Artificial Intelligence, AI 2003, Perth, Australia, 03/12/3. https://doi.org/10.1007/978-3-540-24581-0_84

BN+BN : Behavior network with bayesian network for intelligent agent. / Kim, Kyung Joong; Cho, Sung Bae.

AI 2003: Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings. ed. / Tamas D. Gedeon; Lance Chun Che Fung; Tamas D. Gedeon. Springer Verlag, 2003. p. 979-991 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2903).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - BN+BN

T2 - 16th Australian Conference on Artificial Intelligence, AI 2003

AU - Kim, Kyung Joong

AU - Cho, Sung Bae

PY - 2003

Y1 - 2003

N2 - In the philosophy of behavior-based robotics, design of complex behavior needs the interaction of basic behaviors that are easily implemented. Action selection mechanism selects the most appropriate behavior among them to achieve goals of robot. Usually, robot might have one or more goals that conflict each other and needs a mechanism to coordinate them. Bayesian network represents the dependencies among variables with directed acyclic graph and infers posterior probability using prior knowledge. This paper proposes a method to improve behavior network, action selection mechanism that uses the graph of behaviors, goals and sensors with activation spreading, using goal inference mechanism of Bayesian network learned automatically. Experimental results on Khepera mobile robot show that the proposed method can generate more appropriate behaviors.

AB - In the philosophy of behavior-based robotics, design of complex behavior needs the interaction of basic behaviors that are easily implemented. Action selection mechanism selects the most appropriate behavior among them to achieve goals of robot. Usually, robot might have one or more goals that conflict each other and needs a mechanism to coordinate them. Bayesian network represents the dependencies among variables with directed acyclic graph and infers posterior probability using prior knowledge. This paper proposes a method to improve behavior network, action selection mechanism that uses the graph of behaviors, goals and sensors with activation spreading, using goal inference mechanism of Bayesian network learned automatically. Experimental results on Khepera mobile robot show that the proposed method can generate more appropriate behaviors.

UR - http://www.scopus.com/inward/record.url?scp=7444239161&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=7444239161&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-24581-0_84

DO - 10.1007/978-3-540-24581-0_84

M3 - Conference contribution

AN - SCOPUS:7444239161

SN - 9783540206460

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 979

EP - 991

BT - AI 2003

A2 - Gedeon, Tamas D.

A2 - Fung, Lance Chun Che

A2 - Gedeon, Tamas D.

PB - Springer Verlag

Y2 - 3 December 2003 through 5 December 2003

ER -

Kim KJ, Cho SB. BN+BN: Behavior network with bayesian network for intelligent agent. In Gedeon TD, Fung LCC, Gedeon TD, editors, AI 2003: Advances in Artificial Intelligence - 16th Australian Conference on AI, Proceedings. Springer Verlag. 2003. p. 979-991. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-540-24581-0_84

BN+BN: Behavior network with bayesian network for intelligent agent

Abstract

Publication series

Other

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this