Abstract>
This paper proposes a situation-based conflict estimation method that efficiently generates quality plans for multi-agent systems (MAS) by appropriately selecting abstract plans in hierarchical planning (HP). In HP, selecting a plan at an abstract level affects planning performance because an abstract plan restricts the scope of concrete-level (or primitive) plans and thus can reduce the planning cost. However, if all primitive plans under the selected abstract plan have serious and difficult-to-resolve conflicts with the plans of other agents, the final plan after conflict resolution will be inefficient or of low quality. This issue originates in the uncertainty of MAS, where other agents also have individual plans for their own goals and it is difficult to clearly anticipate which abstract plan will cause fewer conflicts with other agents’ plans. In the proposed method, by introducing conflict patterns that express the situations of conflicts among agents’ plans, agents learn and estimate which abstract plans are less likely to cause conflicts or which conflicts will be easy to resolve; thus, after conflict resolution, they can induce probabilistically higher-utility primitive plans. This paper also describes an experiment to evaluate our method. The results indicate that our method can improve the efficiency of plan execution.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
D. W. Aha, M. Molineaux, and M. Ponsen. Learning to win: Case-based plan selection in a real-time strategy game. Proc. of ICCBR 2005, pages 5–20, 2005.
B. J. Clement, A. C. Barrett, G. R. Rabideau, and E. H. Durfee. Using abstraction in planning and scheduling. Proc. of 6th European Conf. on Planning, 2001.
K. Decker and V. Lesser. Generalizing the Partial Global lanning Algorithm. Int. Jour. on Intelligent Cooperative Information Systems, 1(2):319 – 346, 1992.
T. G. Dietterich. The MAXQ Method for Hierarchical Reinforcement Learning. Proc. of ICML-98, pages 118–126, 1998.
J. A. Giampapa and K. Sycara. Conversational case-based planning for agent team coordination. Proc. of ICCBR 2001, LNAI 2080, pages 189–203, 2001.
R. Goldwin and R. Simmons. Search Control of Plan Generation in Decision- Theoretic Planners. AIPS 1998, pages 94–101, 1998.
J. H. K. Erol and D. S. Nau. HTN planning: Complexity and expressivity. Proc. of AAAI-94, pages 1123–1128, 1994.
L. P. Kaelbling. Hierarchical Learning in Stochastic Domains: Preliminary Results. Proc. of Int. Conf. on Machine Learning, pages 167–173, 1993.
S. Kurihara, S. Aoyagi, T. Takada, T. Hirotsu, and T. Sugawara. Agent-Based Human-Environment Interaction Framework for Ubiquitous Environment. Proc. of Int. Workshop on Networked Sensing Systems, pages 103 – 108, 2005.
L. Macedo and A. Cardoso. Case-Based, Decision-Theoretic, HTN Planning.Proc. of ECCBR 2004, LNAI 3155, pages 257 – 271. Springer-Verlag, 2004.
E. Plaza. Cooperative reuse for compositional cases in multi-agent systems. Proc. of ICCBR 2005, LNAI 3620, pages 382–396, 2005.
E. Sacerdoti. Planning in a hierarchy of abstraction spaces. Artificial Intelligence, 5(2):115 – 135, 1974.
T. Sugawara. Reusing Past Plans in Distributed Planning. Proc. of the 1st Int. Conf. on Multi-Agent Systems (ICMAS95), pages 360 – 367, 1995.
T. Sugawara, S. Kurihara, T. Hirotsu, K. Fukuda, and T. Takada. Predicting Possible Conflicts in Hierarchical planning for Multi-Agent Systems. Proc. of 4th Int. Conf. on Auton. Agents and Multiagent Systems, pages 813 – 820, 2005.
R. S. Sutton, D. Precup, and S. Singh. Intra-Option Learning about Temporary Abstract Actions. Proc. of Int. Conf. on Machine Learning (ICML), pages 556–564, 1998.
T. Takada, S. Kurihara, T. Hirotsu, and T. Sugawara. Proximity Mining: Finding Proximity using Sensor Data History. Proc. of IEEE Workshop on Mobile Computing Systems and Applications, pages 129 – 138, 2003.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag London Limited
About this paper
Cite this paper
Sugawara, T., Kurihara, S., Hirotsu, T., Fukuda, K., Takada, T. (2008). Effective Selection of Abstract Plans for Multi-Agent Systems. In: Bramer, M., Coenen, F., Petridis, M. (eds) Research and Development in Intelligent Systems XXIV. SGAI 2007. Springer, London. https://doi.org/10.1007/978-1-84800-094-0_17
Download citation
DOI: https://doi.org/10.1007/978-1-84800-094-0_17
Publisher Name: Springer, London
Print ISBN: 978-1-84800-093-3
Online ISBN: 978-1-84800-094-0
eBook Packages: Computer ScienceComputer Science (R0)