Grounding Planning Tasks Using Tree Decompositions and Iterated Solving

Corrêa, Augusto; Hecher, Markus; Helmert, Malte; Longo, Davide Mario; Pommerening, Florian; Woltran, Stefan

doi:10.1609/icaps.v33i1.27184

Record link:

http://hdl.handle.net/20.500.12708/192683

Title:

Grounding Planning Tasks Using Tree Decompositions and Iterated Solving

Citation:

Corrêa, A., Hecher, M., Helmert, M., Longo, D. M., Pommerening, F., & Woltran, S. (2023). Grounding Planning Tasks Using Tree Decompositions and Iterated Solving. In Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling (pp. 100–108). Association for the Advancement of Artificial Intelligence. https://doi.org/10.1609/icaps.v33i1.27184

CatalogPlus:

AC17205161

Publisher DOI:

10.1609/icaps.v33i1.27184

Publication Type:

Inproceedings - Full-Paper Contribution

Language:

English

Authors:

Corrêa, Augusto
Hecher, Markus
Helmert, Malte
Longo, Davide Mario
Pommerening, Florian
Woltran, Stefan

Organisational Unit:

E192-02 - Forschungsbereich Databases and Artificial Intelligence

Published in:

Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling

ISBN:

978-1-57735-881-7

Volume:

33,1

Date (published):

1-Jul-2023

Event name:

ICAPS 2023, the 33rd International Conference on Automated Planning and Scheduling

Event date:

8-Jul-2023 - 13-Jul-2023

Event place:

Prague, Czechia

Number of Pages:

Publisher:

Association for the Advancement of Artificial Intelligence, Palo Alto, California, USA

Peer reviewed:

Yes

Keywords:

Tree Decompositions; first-order language; Grounding Planning; Bottleneck; Datalog; recent progress; Baseline

Abstract:

Classical planning tasks are commonly described in a first-order language. However, most classical planners translate tasks by grounding them and then rewriting them into a propositional language. In recent years, the grounding step has become a larger bottleneck. In this work, we study how to improve it. We build on top of the most common grounder for planning tasks which uses Datalog to find all reachable atoms and actions. Inspired by recent progress in lifted planning, database theory, and algorithmics, we develop a new method to ground these Datalog programs. Our algorithm can ground more instances than the baseline, and most tasks it cannot ground are out of reach from any ground planner.

Project title:

Hybrid Parameterized Problem Solving in Practice: P32830-N (FWF - Österr. Wissenschaftsfonds)
Revealing and Utilizing the Hidden Structure for Solving Hard Problems in AI: ICT19-065 (WWTF Wiener Wissenschafts-, Forschu und Technologiefonds)
HyperTrac: hypergraph Decompositions and Tractability: P30930-N35 (FWF - Österr. Wissenschaftsfonds)
Doktoratskolleg: W 1255-N23 (FWF - Österr. Wissenschaftsfonds)

Project (external):

European Union’s Horizon 2020
Austrian Science Fund (FWF)
Gesellschaft für Forschungsförderung Niederösterreich m.b.H. (GFF)

Project ID:

952215
J 4656
ExzF-0004

Research Areas:

Logic and Computation: 100%

Science Branch:

1020 - Informatik: 80%
1010 - Mathematik: 20%

License:

In Copyright

Appears in Collections:

Conference Paper