Extending landmarks analysis to reason about resources and repetition

J. Porteous, S. Cresswell

Research output: Contribution to conferencePaper

Abstract

The identification of important planning subgoals, referred to as landmarks, has been shown to be useful in speeding up plan generation in a number of different planning architec- tures [8]. Here, we report on work in progress to extend some features of this domain analysis using landmarks. In particular we consider: the extraction of resource abstracted landmarks; and the identification of landmark repetition along with a count of the minimum number of times a landmark will need to be repeated. We illustrate our ideas with an example from a benchmark domain and consider ways in which this information could be exploited at plan time.

Conference

ConferenceProceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02)
CityDelft, The Netherlands
Period21/11/0222/11/02

Fingerprint

landmarks
repetition
resources
planning

Keywords

  • planning technology
  • landmarks
  • domain analysis
  • artificial intelligence

Cite this

Porteous, J., & Cresswell, S. (2002). Extending landmarks analysis to reason about resources and repetition. Paper presented at Proceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02), Delft, The Netherlands, .
Porteous, J. ; Cresswell, S. / Extending landmarks analysis to reason about resources and repetition. Paper presented at Proceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02), Delft, The Netherlands, .10 p.
@conference{0ed608008b0d4734a691094371efa672,
title = "Extending landmarks analysis to reason about resources and repetition",
abstract = "The identification of important planning subgoals, referred to as landmarks, has been shown to be useful in speeding up plan generation in a number of different planning architec- tures [8]. Here, we report on work in progress to extend some features of this domain analysis using landmarks. In particular we consider: the extraction of resource abstracted landmarks; and the identification of landmark repetition along with a count of the minimum number of times a landmark will need to be repeated. We illustrate our ideas with an example from a benchmark domain and consider ways in which this information could be exploited at plan time.",
keywords = "planning technology, landmarks, domain analysis, artificial intelligence",
author = "J. Porteous and S. Cresswell",
year = "2002",
month = "11",
language = "English",
note = "Proceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02) ; Conference date: 21-11-2002 Through 22-11-2002",

}

Porteous, J & Cresswell, S 2002, 'Extending landmarks analysis to reason about resources and repetition' Paper presented at Proceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02), Delft, The Netherlands, 21/11/02 - 22/11/02, .

Extending landmarks analysis to reason about resources and repetition. / Porteous, J.; Cresswell, S.

2002. Paper presented at Proceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02), Delft, The Netherlands, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Extending landmarks analysis to reason about resources and repetition

AU - Porteous, J.

AU - Cresswell, S.

PY - 2002/11

Y1 - 2002/11

N2 - The identification of important planning subgoals, referred to as landmarks, has been shown to be useful in speeding up plan generation in a number of different planning architec- tures [8]. Here, we report on work in progress to extend some features of this domain analysis using landmarks. In particular we consider: the extraction of resource abstracted landmarks; and the identification of landmark repetition along with a count of the minimum number of times a landmark will need to be repeated. We illustrate our ideas with an example from a benchmark domain and consider ways in which this information could be exploited at plan time.

AB - The identification of important planning subgoals, referred to as landmarks, has been shown to be useful in speeding up plan generation in a number of different planning architec- tures [8]. Here, we report on work in progress to extend some features of this domain analysis using landmarks. In particular we consider: the extraction of resource abstracted landmarks; and the identification of landmark repetition along with a count of the minimum number of times a landmark will need to be repeated. We illustrate our ideas with an example from a benchmark domain and consider ways in which this information could be exploited at plan time.

KW - planning technology

KW - landmarks

KW - domain analysis

KW - artificial intelligence

UR - http://www.cis.strath.ac.uk/research/publications/papers/strath_cis_publication_81.pdf

M3 - Paper

ER -

Porteous J, Cresswell S. Extending landmarks analysis to reason about resources and repetition. 2002. Paper presented at Proceedings of the 21st Workshop of the UK Planning and Scheduling Special Interest Group (PLANSIG '02), Delft, The Netherlands, .