Planning with Partial Observability by SAT

Saurabh Fadnis; Jussi Rintanen

doi:10.1007/978-3-031-43619-2_41

Planning with Partial Observability by SAT

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

1 Citation (Scopus)

18 Downloads (Pure)

Abstract

Geffner & Geffner (2018) have shown that finding plans by reduction to SAT is not limited to classical planning, but is competitive also for fully observable non-deterministic planning. This work extends these ideas to planning with partial observability. Specifically, we handle partial observability by requiring that during the execution of a plan, the same actions have to be taken in all indistinguishable circumstances. We demonstrate that encoding this condition directly leads to far better scalability than an explicit encoding of observations-to-actions mapping, for high numbers of observations.

Original language	English
Title of host publication	Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings
Editors	Sarah Gaggl, Maria Vanina Martinez, Magdalena Ortiz, Magdalena Ortiz
Publisher	Springer
Pages	605-620
Number of pages	16
ISBN (Print)	978-3-031-43618-5
DOIs	https://doi.org/10.1007/978-3-031-43619-2_41
Publication status	Published - 2023
MoE publication type	A4 Conference publication
Event	European Conference on Logics in Artificial Intelligence - Dresden, Germany Duration: 20 Sept 2023 → 22 Sept 2023 Conference number: 18

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer
Volume	14281 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	European Conference on Logics in Artificial Intelligence
Abbreviated title	JELIA
Country/Territory	Germany
City	Dresden
Period	20/09/2023 → 22/09/2023

Access to Document

10.1007/978-3-031-43619-2_41

Planning with Partial Observability by SATAccepted author manuscript, 317 KB

Cite this

Fadnis, S., & Rintanen, J. (2023). Planning with Partial Observability by SAT. In S. Gaggl, M. V. Martinez, M. Ortiz, & M. Ortiz (Eds.), Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings (pp. 605-620). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14281 LNAI). Springer. https://doi.org/10.1007/978-3-031-43619-2_41

Fadnis, Saurabh ; Rintanen, Jussi. / Planning with Partial Observability by SAT. Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings. editor / Sarah Gaggl ; Maria Vanina Martinez ; Magdalena Ortiz ; Magdalena Ortiz. Springer, 2023. pp. 605-620 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{09adc9affcec4951b3c97f70afbd82e8,

title = "Planning with Partial Observability by SAT",

abstract = "Geffner & Geffner (2018) have shown that finding plans by reduction to SAT is not limited to classical planning, but is competitive also for fully observable non-deterministic planning. This work extends these ideas to planning with partial observability. Specifically, we handle partial observability by requiring that during the execution of a plan, the same actions have to be taken in all indistinguishable circumstances. We demonstrate that encoding this condition directly leads to far better scalability than an explicit encoding of observations-to-actions mapping, for high numbers of observations.",

author = "Saurabh Fadnis and Jussi Rintanen",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.; European Conference on Logics in Artificial Intelligence, JELIA ; Conference date: 20-09-2023 Through 22-09-2023",

year = "2023",

doi = "10.1007/978-3-031-43619-2_41",

language = "English",

isbn = "978-3-031-43618-5",

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

publisher = "Springer",

pages = "605--620",

editor = "Sarah Gaggl and Martinez, {Maria Vanina} and Magdalena Ortiz and Magdalena Ortiz",

booktitle = "Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings",

address = "Germany",

}

Fadnis, S & Rintanen, J 2023, Planning with Partial Observability by SAT. in S Gaggl, MV Martinez, M Ortiz & M Ortiz (eds), Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 14281 LNAI, Springer, pp. 605-620, European Conference on Logics in Artificial Intelligence, Dresden, Germany, 20/09/2023. https://doi.org/10.1007/978-3-031-43619-2_41

Planning with Partial Observability by SAT. / Fadnis, Saurabh; Rintanen, Jussi.
Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings. ed. / Sarah Gaggl; Maria Vanina Martinez; Magdalena Ortiz; Magdalena Ortiz. Springer, 2023. p. 605-620 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 14281 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Planning with Partial Observability by SAT

AU - Fadnis, Saurabh

AU - Rintanen, Jussi

N1 - Conference code: 18

PY - 2023

Y1 - 2023

N2 - Geffner & Geffner (2018) have shown that finding plans by reduction to SAT is not limited to classical planning, but is competitive also for fully observable non-deterministic planning. This work extends these ideas to planning with partial observability. Specifically, we handle partial observability by requiring that during the execution of a plan, the same actions have to be taken in all indistinguishable circumstances. We demonstrate that encoding this condition directly leads to far better scalability than an explicit encoding of observations-to-actions mapping, for high numbers of observations.

AB - Geffner & Geffner (2018) have shown that finding plans by reduction to SAT is not limited to classical planning, but is competitive also for fully observable non-deterministic planning. This work extends these ideas to planning with partial observability. Specifically, we handle partial observability by requiring that during the execution of a plan, the same actions have to be taken in all indistinguishable circumstances. We demonstrate that encoding this condition directly leads to far better scalability than an explicit encoding of observations-to-actions mapping, for high numbers of observations.

UR - http://www.scopus.com/inward/record.url?scp=85174540813&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-43619-2_41

DO - 10.1007/978-3-031-43619-2_41

M3 - Conference article in proceedings

AN - SCOPUS:85174540813

SN - 978-3-031-43618-5

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

SP - 605

EP - 620

BT - Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings

A2 - Gaggl, Sarah

A2 - Martinez, Maria Vanina

A2 - Ortiz, Magdalena

PB - Springer

T2 - European Conference on Logics in Artificial Intelligence

Y2 - 20 September 2023 through 22 September 2023

ER -

Fadnis S, Rintanen J. Planning with Partial Observability by SAT. In Gaggl S, Martinez MV, Ortiz M, Ortiz M, editors, Logics in Artificial Intelligence - 18th European Conference, JELIA 2023, Proceedings. Springer. 2023. p. 605-620. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-43619-2_41