Abstract
In this work, we explore how local interactions can simplify the process of decision-making in multiagent systems, particularly in multirobot problems. We review a recent decision-theoretic model for multiagent systems, the decentralized sparse-interaction Markov decision process (Dec-SIMDP), that explicitly distinguishes the situations in which the agents in the team must coordinate from those in which they can act independently. We situate this class of problems within different multiagent models, such as MMDPs and transition independent Dec-MDPs. We then contribute a new general approach that leverages the particular structure of Dec-SIMDPs to efficiently plan in this class of problems, and propose two algorithms based on this underlying approach. We pinpoint the main properties of our approach through illustrative examples in multirobot navigation domains with partial observability, and provide empirical comparisons between our algorithms and other existing algorithms for this class of problems. We show that our approach allows the robots to look ahead for possible interactions, planning to accommodate such interactions and thus overcome some of the limitations of previous methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Allen, M., Zilberstein, S.: Complexity of Decentralized Control: Special Cases. In: Adv. Neural. Information Proc. Systems, pp. 19–27 (2009)
Becker, R., Zilberstein, S., Lesser, V., Goldman, C.: Solving transition independent decentralized Markov decision processes. J. Artif. Intell. Res. 22, 423–455 (2004)
Bernstein, D., Givan, R., Immerman, N., Zilberstein, S.: The complexity of decentralized control of Markov decision processes. Math. Oper. Res. 27(4), 819–840 (2002)
Bernstein, D., Amato, C., Zilberstein, S.: Policy iteration for decentralized control of Markov decision processes. J. Artif. Intell. Res. 34, 89–132 (2009)
Gerkey, B., Matarić, M.: Sold!: Auction methods for multirobot coordination. IEEE T. Robot. Autom. 18(5), 758–768 (2002)
Ghavamzadeh, M., Mahadevan, S., Makar, R.: Hierarchical multiagent reinforcement learning. J. Auton. Agent Multiag. 13(2), 197–229 (2006)
Goldman, C., Zilberstein, S.: Decentralized control of cooperative systems: Categorization and complexity analysis. J. Artif. Intell. Res. 22, 143–174 (2004)
Guestrin, C., Koller, D., Parr, R.: Multiagent planning with factored MDPs. In: Adv. Neural Information Proc. Systems, pp. 1523–1530 (2001)
Kearns, M., Littman, M., Singh, S.: Graphical models for game theory. In: Conf. Uncert. Artif. Intell., pp. 253–260 (2001)
Kok, J., Hoen, P., Bakker, B., Vlassis, N.: Utile coordination: Learning interdependencies among cooperative agents. In: IEEE Symp. Comput. Intell. Games, pp. 61–68 (2005)
Littman, M., Cassandra, A., Kaelbling, L.: Learning policies for partially observable environments: Scaling up. In: Int. Conf. Mach. Learn., pp. 362–370 (1995)
Madani, O., Hanks, S., Condon, A.: On the undecidability of probabilistic planning in infinite-horizon partially observable Markov decision problems. In: AAAI Conf. Artif. Intell., pp. 541–548 (1999)
Melo, F., Veloso, M.: Local Multiagent Coordination in Decentralized MDPs with Sparse Interactions. Tech. Report CMU-CS-10-133, CS Dep., Carnegie Mellon Univ. (2010)
Mostafa, H., Lesser, V.: Offline planning for communication by exploiting structured interactions in decentralized MDPs. Tech Rep. TR 2009-020, CS Dep., Univ. Massachusetts (2009)
Parker, L.: ALLIANCE: An architecture for fault-tolerant multirobot cooperation. IEEE T. Robot. Autom. 14(2), 220–240 (1998)
Roth, M., Simmons, R., Veloso, M.: Exploiting factored representations for decentralized execution in multiagent teams. In: Int. Conf. Auton. Agent Multiag., pp. 469–475 (2007)
Spaan, M., Melo, F.: Interaction-driven Markov games for decentralized multiagent planning under uncertainty. In: Int. Conf. Auton. Agent Multiag., pp. 525–532 (2008)
Stone, P.: Layered learning in multiagent systems. PhD thesis, Carnegie Mellon Univ. (1998)
Varakantham, P., Kwak, J., Taylor, M., Marecki, J., Scerri, P., Tambe, M.: Exploiting coordination locales in distributed POMDPs via social model shaping. In: Int. Conf. Autom. Plan Scheduling, pp. 313–320 (2009)
Xin Jiang, A., Leyton-Brown, K., Bhat, N.: Action-graph games. Tech Rep. TR-2008-13, Univ. British Columbia (2008)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Melo, F.S., Veloso, M. (2013). Heuristic Planning for Decentralized MDPs with Sparse Interactions. In: Martinoli, A., et al. Distributed Autonomous Robotic Systems. Springer Tracts in Advanced Robotics, vol 83. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32723-0_24
Download citation
DOI: https://doi.org/10.1007/978-3-642-32723-0_24
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32722-3
Online ISBN: 978-3-642-32723-0
eBook Packages: EngineeringEngineering (R0)