Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications

Francesco Percassi; Enrico Scala; Mauro Vallati

doi:10.1609/icaps.v33i1.27210

Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications

Francesco Percassi, Enrico Scala, Mauro Vallati

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

3 Citations (Scopus)

Abstract

The plan, execution, and replan framework has proven to be extremely valuable in complex real-world applications, where the dynamics of the environment cannot be fully encoded in the domain model. However, this comes at the cost of regenerating plans from scratch, which can be expensive when expressive formalisms like PDDL+ are used. Given the complexity of generating PDDL+ plans, it would be ideal to reuse as much as possible of an existing plan, rather than generating a new one from scratch every time. To support more effective exploitation of the plan, execution, and replan framework in PDDL+, in this paper, we introduce the problem of discretised PDDL+ plan fixing, which allows one to fix existing plans according to some defined constraints. We demonstrate the theoretical implications of the introduced notion and introduce reformulations to address the problem using domain-independent planning engines. Our results show that such reformulations can outperform replanning from scratch and unlock planning engines to solve more problems with fine-grained discretisations.

Original language	English
Title of host publication	Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling
Subtitle of host publication	(ICAPS 2023)
Editors	Sven Koenig, Roni Stern, Mauro Vallati
Publisher	AAAI press
Pages	324-333
Number of pages	10
Volume	33
Edition	1
ISBN (Print)	1577358813, 9781577358817
DOIs	https://doi.org/10.1609/icaps.v33i1.27210
Publication status	Published - 2 Jul 2023
Event	33rd International Conference on Automated Planning and Scheduling - Charles University, Prague, Czech Republic Duration: 8 Jul 2023 → 13 Jul 2023 Conference number: 33 https://icaps23.icaps-conference.org/

Publication series

Name	Proceedings of the International Conference on Automated Planning and Scheduling
Publisher	AAAI Press
Number	1
Volume	33
ISSN (Print)	2334-0835
ISSN (Electronic)	2334-0843

Conference

Conference	33rd International Conference on Automated Planning and Scheduling
Abbreviated title	ICAPS 2023
Country/Territory	Czech Republic
City	Prague
Period	8/07/23 → 13/07/23
Internet address	https://icaps23.icaps-conference.org/

Access to Document

10.1609/icaps.v33i1.27210Licence: Unspecified

Cite this

Percassi, F., Scala, E., & Vallati, M. (2023). Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications. In S. Koenig, R. Stern, & M. Vallati (Eds.), Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling: (ICAPS 2023) (1 ed., Vol. 33, pp. 324-333). (Proceedings of the International Conference on Automated Planning and Scheduling; Vol. 33, No. 1). AAAI press. https://doi.org/10.1609/icaps.v33i1.27210

Percassi, Francesco ; Scala, Enrico ; Vallati, Mauro. / Fixing Plans for PDDL+ Problems : Theoretical and Practical Implications. Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling: (ICAPS 2023). editor / Sven Koenig ; Roni Stern ; Mauro Vallati. Vol. 33 1. ed. AAAI press, 2023. pp. 324-333 (Proceedings of the International Conference on Automated Planning and Scheduling; 1).

@inproceedings{d9bb86aed2414838b06f06636b2b3a20,

title = "Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications",

abstract = "The plan, execution, and replan framework has proven to be extremely valuable in complex real-world applications, where the dynamics of the environment cannot be fully encoded in the domain model. However, this comes at the cost of regenerating plans from scratch, which can be expensive when expressive formalisms like PDDL+ are used. Given the complexity of generating PDDL+ plans, it would be ideal to reuse as much as possible of an existing plan, rather than generating a new one from scratch every time. To support more effective exploitation of the plan, execution, and replan framework in PDDL+, in this paper, we introduce the problem of discretised PDDL+ plan fixing, which allows one to fix existing plans according to some defined constraints. We demonstrate the theoretical implications of the introduced notion and introduce reformulations to address the problem using domain-independent planning engines. Our results show that such reformulations can outperform replanning from scratch and unlock planning engines to solve more problems with fine-grained discretisations.",

keywords = "Plan and schedule execution, monitoring, and repair, Planning and scheduling with mixed continuous and discrete states/actions/decisions, Knowledge representation for planning and scheduling",

author = "Francesco Percassi and Enrico Scala and Mauro Vallati",

note = "Funding Information: Francesco Percassi and Mauro Vallati were supported by a UKRI Future Leaders Fellowship [grant number MR/T041196/1]. Enrico Scala has been partially supported by AIPlan4EU, a project funded by EU Horizon 2020 research and innovation programme under GA n. 101016442. Publisher Copyright: Copyright {\textcopyright} 2023, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.; 33rd International Conference on Automated Planning and Scheduling , ICAPS 2023 ; Conference date: 08-07-2023 Through 13-07-2023",

year = "2023",

month = jul,

day = "2",

doi = "10.1609/icaps.v33i1.27210",

language = "English",

isbn = "1577358813",

volume = "33",

series = "Proceedings of the International Conference on Automated Planning and Scheduling",

publisher = "AAAI press",

number = "1",

pages = "324--333",

editor = "Sven Koenig and Roni Stern and Mauro Vallati",

booktitle = "Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling",

address = "United States",

edition = "1",

url = "https://icaps23.icaps-conference.org/",

}

Percassi, F, Scala, E & Vallati, M 2023, Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications. in S Koenig, R Stern & M Vallati (eds), Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling: (ICAPS 2023). 1 edn, vol. 33, Proceedings of the International Conference on Automated Planning and Scheduling, no. 1, vol. 33, AAAI press, pp. 324-333, 33rd International Conference on Automated Planning and Scheduling , Prague, Czech Republic, 8/07/23. https://doi.org/10.1609/icaps.v33i1.27210

Fixing Plans for PDDL+ Problems : Theoretical and Practical Implications. / Percassi, Francesco; Scala, Enrico; Vallati, Mauro.

Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling: (ICAPS 2023). ed. / Sven Koenig; Roni Stern; Mauro Vallati. Vol. 33 1. ed. AAAI press, 2023. p. 324-333 (Proceedings of the International Conference on Automated Planning and Scheduling; Vol. 33, No. 1).

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

TY - GEN

T1 - Fixing Plans for PDDL+ Problems

T2 - 33rd International Conference on Automated Planning and Scheduling

AU - Percassi, Francesco

AU - Scala, Enrico

AU - Vallati, Mauro

N1 - Conference code: 33

PY - 2023/7/2

Y1 - 2023/7/2

N2 - The plan, execution, and replan framework has proven to be extremely valuable in complex real-world applications, where the dynamics of the environment cannot be fully encoded in the domain model. However, this comes at the cost of regenerating plans from scratch, which can be expensive when expressive formalisms like PDDL+ are used. Given the complexity of generating PDDL+ plans, it would be ideal to reuse as much as possible of an existing plan, rather than generating a new one from scratch every time. To support more effective exploitation of the plan, execution, and replan framework in PDDL+, in this paper, we introduce the problem of discretised PDDL+ plan fixing, which allows one to fix existing plans according to some defined constraints. We demonstrate the theoretical implications of the introduced notion and introduce reformulations to address the problem using domain-independent planning engines. Our results show that such reformulations can outperform replanning from scratch and unlock planning engines to solve more problems with fine-grained discretisations.

AB - The plan, execution, and replan framework has proven to be extremely valuable in complex real-world applications, where the dynamics of the environment cannot be fully encoded in the domain model. However, this comes at the cost of regenerating plans from scratch, which can be expensive when expressive formalisms like PDDL+ are used. Given the complexity of generating PDDL+ plans, it would be ideal to reuse as much as possible of an existing plan, rather than generating a new one from scratch every time. To support more effective exploitation of the plan, execution, and replan framework in PDDL+, in this paper, we introduce the problem of discretised PDDL+ plan fixing, which allows one to fix existing plans according to some defined constraints. We demonstrate the theoretical implications of the introduced notion and introduce reformulations to address the problem using domain-independent planning engines. Our results show that such reformulations can outperform replanning from scratch and unlock planning engines to solve more problems with fine-grained discretisations.

KW - Plan and schedule execution, monitoring, and repair

KW - Planning and scheduling with mixed continuous and discrete states/actions/decisions

KW - Knowledge representation for planning and scheduling

UR - http://www.scopus.com/inward/record.url?scp=85169780142&partnerID=8YFLogxK

U2 - 10.1609/icaps.v33i1.27210

DO - 10.1609/icaps.v33i1.27210

M3 - Conference contribution

SN - 1577358813

SN - 9781577358817

VL - 33

T3 - Proceedings of the International Conference on Automated Planning and Scheduling

SP - 324

EP - 333

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

A2 - Koenig, Sven

A2 - Stern, Roni

A2 - Vallati, Mauro

PB - AAAI press

Y2 - 8 July 2023 through 13 July 2023

ER -

Percassi F, Scala E, Vallati M. Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications. In Koenig S, Stern R, Vallati M, editors, Proceedings of the Thirty-Third International Conference on Automated Planning and Scheduling: (ICAPS 2023). 1 ed. Vol. 33. AAAI press. 2023. p. 324-333. (Proceedings of the International Conference on Automated Planning and Scheduling; 1). Epub 2023 Jul 1. doi: 10.1609/icaps.v33i1.27210

Fixing Plans for PDDL+ Problems: Theoretical and Practical Implications

Abstract

Publication series

Conference

Access to Document

Other files and links

Cite this