Hostname: page-component-8448b6f56d-mp689 Total loading time: 0 Render date: 2024-04-23T15:16:12.971Z Has data issue: false hasContentIssue false
  • English
  • Français

Acquiring planning domain models using LOCM

Published online by Cambridge University Press:  22 February 2013

Stephen N. Cresswell ,
Thomas L. McCluskey  and
Margaret M. West
Stephen N. Cresswell
Affiliation:
The Stationery Office, St. Crispins, Duke Street, Norwich NR3 1PD, UK; e-mail: stephen.cresswell@tso.co.uk
Thomas L. McCluskey
Affiliation:
School of Computing and Engineering, The University of Huddersfield, Huddersfield HD1 3DH, UK; e-mail: t.l.mccluskey@hud.ac.uk, m.m.west@hud.ac.uk
Margaret M. West
Affiliation:
School of Computing and Engineering, The University of Huddersfield, Huddersfield HD1 3DH, UK; e-mail: t.l.mccluskey@hud.ac.uk, m.m.west@hud.ac.uk
Get access Rights & Permissions [Opens in a new window]

Abstract

The problem of formulating knowledge bases containing action schema is a central concern in knowledge engineering for artificial intelligence (AI) planning. This paper describes Learning Object-Centred Models (LOCM), a system that carries out the automated generation of a planning domain model from example training plans. The novelty of LOCM is that it can induce action schema without being provided with any information about predicates or initial, goal or intermediate state descriptions for the example action sequences. Each plan is assumed to be a sound sequence of actions; each action in a plan is stated as a name and a list of objects that the action refers to. LOCM exploits assumptions about the kinds of domain model it has to generate, rather than handcrafted clues or planner-oriented knowledge. It assumes that actions change the state of objects, and require objects to be in a certain state before they can be executed. In this paper, we describe the implemented LOCM algorithm, the assumptions that it is based on, and an evaluation using plans generated through goal-directed solutions, through random walk, and through logging human-generated plans for the game of freecell. We analyze the performance of LOCM by its application to the induction of domain models from five domains.

Type
Articles
Information
The Knowledge Engineering Review , Volume 28 , Special Issue 2: Knowledge Engineering for Planning and Scheduling , June 2013 , pp. 195 - 213
Copyright
Copyright © Cambridge University Press 2013 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Bäckström, C. 1992. Equivalence and tractability results for SAS+ planning. In Proceedings of the 3rd International Conference on Principles on Knowledge Representation and Reasoning (KR-92), Swartout, B. & Nebel, B. (eds). Morgan Kaufmann.Google Scholar
Benson, S. S. 1996. Learning Action Models for Reactive Autonomous Agents. PhD dissertation, Department of Computer Science, Stanford University.Google Scholar
Fikes, R., Hart, P. E., Nilsson, N. J. 1972. Learning and executing generalized robot plans. Artificial Intelligence 3(1–3), 251–288.CrossRefGoogle Scholar
Fox, M., Long, D. 1998. The automatic inference of state invariants in TIM. Journal of Artificial Intelligence Research 9, 367–421.CrossRefGoogle Scholar
Grant, T. J. 1996. Inductive Learning of Knowledge-Based Planning Operators. PhD dissertation, de Rijksuniversiteit Limburg te Maastricht, The Netherlands.Google Scholar
Grant, T. J. 2007. Assimilating planning domain knowledge from other agents. In Proceedings of the 26th Workshop of the UK Planning and Scheduling Special Interest Group, Prague, Czech Republic.Google Scholar
Hoffmann, J., Weber, I., Kraft, F. M. 2009. Planning@SAP: an application in business process management. In Proceedings of the 2nd International Scheduling and Planning Applications woRKshop (SPARK'09), at ICAPS'09, Thessaloniki, Greece.Google Scholar
McCluskey, T. L., Richardson, N. E., Simpson, R. M. 2002. An Interactive Method for Inducing Operator Descriptions. In The 6th International Conference on Artificial Intelligence Planning Systems (AIPS), Ghallab, M., Hertzberg, J. & Traverso, P. (eds). Toulouse, France, 151–160. AAAI Press.Google Scholar
McCluskey, T. L., Cresswell, S. N., Richardson, N. E., West, M. M. 2009. Automated acquisition of action knowledge. In International Conference on Agents and Artificial Intelligence (ICAART), Porto, Portugal, 93–100.Google Scholar
Quinlan, J. 1990. Learning logical definitions from relations. Machine Learning 5, 239–266.CrossRefGoogle Scholar
Richardson, N. E. 2008. An Operator Induction Tool Supporting Knowledge Engineering in Planning. PhD dissertation, School of Computing and Engineering, University of Huddersfield, UK.Google Scholar
Shahaf, D., Amir, E. 2006. Learning partially observable action schemas. In Proceedings of the 21st National Conference on Artificial Intelligence and the 18th Innovative Applications of Artificial Intelligence Conference, July 16–20, 2006, Boston, Massachusetts, USA. AAAI Press, http://www.aaai.org/Press/Proceedings/aaai06.php, http://dblp.uni-trier.de/rec/bibtex/conf/aaai/2006.Google Scholar
Simpson, R. M., Kitchin, D. E., McCluskey, T. L. 2007. Planning domain definition using GIPO. Knowledge Engineering Review 22(2), 117–134.CrossRefGoogle Scholar
Wu, K., Yang, Q., Jiang, Y. 2005. ARMS: Action-relation modelling system for learning acquisition models. In Proceedings of the 1st International Competition on Knowledge Engineering for AI Planning, Monterey, California, USA.Google Scholar
Zhuo, H. H., Hu, D. H., Yang, Q. 2009. Learning applicability conditions in AI planning from partial observations. In Workshop on Learning Structural Knowledge From Observations at IJCAI-09, Pasadena, California, USA.Google Scholar