Exploring lifted planning encodings in Essence Prime

Joan Espasa; Jordi Coll; Ian Miguel; Mateu Villaret

doi:10.3233/faia210117

Exploring lifted planning encodings in Essence Prime

Joan Espasa, Jordi Coll, Ian Miguel, Mateu Villaret

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

16 Downloads (Pure)

Abstract

State-space planning is the de-facto search method of the automated planning community. Planning problems are typically expressed in the Planning Domain Definition Language (PDDL), where action and variable templates describe the sets of actions and variables that occur in the problem. Typically, a planner begins by generating the full set of instantiations of these templates, which in turn are used to derive useful heuristics that guide the search. Thanks to this success, there has been limited research in other directions. We explore a different approach, keeping the compact representation by directly reformulating the problem in PDDL into ESSENCE PRIME, a Constraint Programming language with support for distinct solving technologies including SAT and SMT. In particular, we explore two different encodings from PDDL to ESSENCE PRIME, how they represent action parameters, and their performance. The encodings are able to maintain the compactness of the PDDL representation, and while they differ slightly, they perform quite differently on various instances from the International Planning Competition.

Original language	English
Title of host publication	Artificial Intelligence Research and Development
Subtitle of host publication	Proceedings of the 23rd International Conference of the Catalan Association for Artificial Intelligence
Editors	Mateu Villaret, Teresa Alsinet, Cèsar Fernández, Aïda Valls
Publisher	IOS Press
Pages	66-75
ISBN (Electronic)	9781643682112
ISBN (Print)	9781643682105
DOIs	https://doi.org/10.3233/faia210117
Publication status	Published - 14 Oct 2021
Event	23rd International Conference of the Catalan Association for Artificial Intelligence - University of Lleida Duration: 20 Oct 2021 → 22 Oct 2021 Conference number: 23 https://ccia2021.udl.cat/en/english/

Publication series

Name	Frontiers in Artificial Intelligence and Applications
Publisher	IOS Press
Volume	339
ISSN (Print)	0922-6389

Conference

Conference	23rd International Conference of the Catalan Association for Artificial Intelligence
Abbreviated title	CCIA
Period	20/10/21 → 22/10/21
Internet address	https://ccia2021.udl.cat/en/english/

Keywords

Reformulation
Modelling
Automated Planning
Constraint Programming

Access to Document

10.3233/faia210117Licence: CC BY-NC

Espasa_2021_AIRD_Exploring-lifted-planning_CC
Copyright © 2021 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
Final published version, 226 KBLicence: CC BY-NC

Keep Learning: Keep Learning
Miguel, I. J. (PI)
EPSRC
1/07/21 → 31/12/24
Project: Standard

Cite this

Espasa, J., Coll, J., Miguel, I., & Villaret, M. (2021). Exploring lifted planning encodings in Essence Prime. In M. Villaret, T. Alsinet, C. Fernández, & A. Valls (Eds.), Artificial Intelligence Research and Development: Proceedings of the 23rd International Conference of the Catalan Association for Artificial Intelligence (pp. 66-75). (Frontiers in Artificial Intelligence and Applications; Vol. 339). IOS Press. https://doi.org/10.3233/faia210117

Espasa, Joan ; Coll, Jordi ; Miguel, Ian et al. / Exploring lifted planning encodings in Essence Prime. Artificial Intelligence Research and Development: Proceedings of the 23rd International Conference of the Catalan Association for Artificial Intelligence. editor / Mateu Villaret ; Teresa Alsinet ; Cèsar Fernández ; Aïda Valls. IOS Press, 2021. pp. 66-75 (Frontiers in Artificial Intelligence and Applications).

@inproceedings{ec4e07fc761846a586af46c884795730,

title = "Exploring lifted planning encodings in Essence Prime",

abstract = "State-space planning is the de-facto search method of the automated planning community. Planning problems are typically expressed in the Planning Domain Definition Language (PDDL), where action and variable templates describe the sets of actions and variables that occur in the problem. Typically, a planner begins by generating the full set of instantiations of these templates, which in turn are used to derive useful heuristics that guide the search. Thanks to this success, there has been limited research in other directions. We explore a different approach, keeping the compact representation by directly reformulating the problem in PDDL into ESSENCE PRIME, a Constraint Programming language with support for distinct solving technologies including SAT and SMT. In particular, we explore two different encodings from PDDL to ESSENCE PRIME, how they represent action parameters, and their performance. The encodings are able to maintain the compactness of the PDDL representation, and while they differ slightly, they perform quite differently on various instances from the International Planning Competition.",

keywords = "Reformulation, Modelling, Automated Planning, Constraint Programming",

author = "Joan Espasa and Jordi Coll and Ian Miguel and Mateu Villaret",

note = "This work is supported by UK EPSRC EP/P015638/1 and EP/V027182/1, by the MICINN/FEDER, UE (RTI2018-095609-B-I00), by the French Agence Nationale de la Recherche, reference ANR-19-CHIA-0013-01, and by Archimedes institute, Aix-Marseille University.; 23rd International Conference of the Catalan Association for Artificial Intelligence, CCIA ; Conference date: 20-10-2021 Through 22-10-2021",

year = "2021",

month = oct,

day = "14",

doi = "10.3233/faia210117",

language = "English",

isbn = "9781643682105",

series = "Frontiers in Artificial Intelligence and Applications",

publisher = "IOS Press",

pages = "66--75",

editor = "Mateu Villaret and Teresa Alsinet and C{\`e}sar Fern{\'a}ndez and A{\"i}da Valls",

booktitle = "Artificial Intelligence Research and Development",

address = "Netherlands",

url = "https://ccia2021.udl.cat/en/english/",

}

Espasa, J, Coll, J, Miguel, I & Villaret, M 2021, Exploring lifted planning encodings in Essence Prime. in M Villaret, T Alsinet, C Fernández & A Valls (eds), Artificial Intelligence Research and Development: Proceedings of the 23rd International Conference of the Catalan Association for Artificial Intelligence. Frontiers in Artificial Intelligence and Applications, vol. 339, IOS Press, pp. 66-75, 23rd International Conference of the Catalan Association for Artificial Intelligence, 20/10/21. https://doi.org/10.3233/faia210117

Exploring lifted planning encodings in Essence Prime. / Espasa, Joan; Coll, Jordi; Miguel, Ian et al.
Artificial Intelligence Research and Development: Proceedings of the 23rd International Conference of the Catalan Association for Artificial Intelligence. ed. / Mateu Villaret; Teresa Alsinet; Cèsar Fernández; Aïda Valls. IOS Press, 2021. p. 66-75 (Frontiers in Artificial Intelligence and Applications; Vol. 339).

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

TY - GEN

T1 - Exploring lifted planning encodings in Essence Prime

AU - Espasa, Joan

AU - Coll, Jordi

AU - Miguel, Ian

AU - Villaret, Mateu

N1 - Conference code: 23

PY - 2021/10/14

Y1 - 2021/10/14

N2 - State-space planning is the de-facto search method of the automated planning community. Planning problems are typically expressed in the Planning Domain Definition Language (PDDL), where action and variable templates describe the sets of actions and variables that occur in the problem. Typically, a planner begins by generating the full set of instantiations of these templates, which in turn are used to derive useful heuristics that guide the search. Thanks to this success, there has been limited research in other directions. We explore a different approach, keeping the compact representation by directly reformulating the problem in PDDL into ESSENCE PRIME, a Constraint Programming language with support for distinct solving technologies including SAT and SMT. In particular, we explore two different encodings from PDDL to ESSENCE PRIME, how they represent action parameters, and their performance. The encodings are able to maintain the compactness of the PDDL representation, and while they differ slightly, they perform quite differently on various instances from the International Planning Competition.

AB - State-space planning is the de-facto search method of the automated planning community. Planning problems are typically expressed in the Planning Domain Definition Language (PDDL), where action and variable templates describe the sets of actions and variables that occur in the problem. Typically, a planner begins by generating the full set of instantiations of these templates, which in turn are used to derive useful heuristics that guide the search. Thanks to this success, there has been limited research in other directions. We explore a different approach, keeping the compact representation by directly reformulating the problem in PDDL into ESSENCE PRIME, a Constraint Programming language with support for distinct solving technologies including SAT and SMT. In particular, we explore two different encodings from PDDL to ESSENCE PRIME, how they represent action parameters, and their performance. The encodings are able to maintain the compactness of the PDDL representation, and while they differ slightly, they perform quite differently on various instances from the International Planning Competition.

KW - Reformulation

KW - Modelling

KW - Automated Planning

KW - Constraint Programming

U2 - 10.3233/faia210117

DO - 10.3233/faia210117

M3 - Conference contribution

SN - 9781643682105

T3 - Frontiers in Artificial Intelligence and Applications

SP - 66

EP - 75

BT - Artificial Intelligence Research and Development

A2 - Villaret, Mateu

A2 - Alsinet, Teresa

A2 - Fernández, Cèsar

A2 - Valls, Aïda

PB - IOS Press

T2 - 23rd International Conference of the Catalan Association for Artificial Intelligence

Y2 - 20 October 2021 through 22 October 2021

ER -

Espasa J, Coll J, Miguel I, Villaret M. Exploring lifted planning encodings in Essence Prime. In Villaret M, Alsinet T, Fernández C, Valls A, editors, Artificial Intelligence Research and Development: Proceedings of the 23rd International Conference of the Catalan Association for Artificial Intelligence. IOS Press. 2021. p. 66-75. (Frontiers in Artificial Intelligence and Applications). doi: 10.3233/faia210117

Exploring lifted planning encodings in Essence Prime

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Projects

Keep Learning: Keep Learning

Cite this