Decomposition-Based Optimal Market-Based Planning for Multi-Agent Systems with Shared Resources

Market-based algorithms have become popular in collaborative multi-agent planning due to their simplicity, distributedness, low communication requirements, and proven success in domains such as task allocation and robotic exploration. Most existing marketbased algorithms, however, suer from two main drawbacks: resource prices must be carefully handcrafted for each problem domain, and there is no guarantee on nal solution quality. We present an optimal marketbased algorithm, derived from a mixed integer program formulation of planning problems. Our method is based on two wellknown techniques for optimization: DantzigWolfe decomposition and Gomory cuts. The former prices resources optimally for a relaxed version of the problem, while the latter introduces new derivative resources to correct pricing imbalances that arise from the relaxation. Our algorithm is applicable to a wide variety of multi-agent planning domains. We provide optimality guarantees and demonstrate the eectiveness of our algorithm in both centralized and distributed settings on synthetic planning problems.

[1]  Geoffrey J. Gordon,et al.  No-regret learning and a mechanism for distributed multiagent planning , 2008, AAMAS.

[2]  George B. Dantzig,et al.  Linear programming and extensions , 1965 .

[3]  Sven Koenig,et al.  Progress on Agent Coordination with Cooperative Auctions , 2010, AAAI.

[4]  Bart Selman,et al.  Satisfiability Solvers , 2008, Handbook of Knowledge Representation.

[5]  Ralph E. Gomory,et al.  Outline of an Algorithm for Integer Solutions to Linear Programs and An Algorithm for the Mixed Integer Problem , 2010, 50 Years of Integer Programming.

[6]  François Vanderbeck,et al.  Implementing Mixed Integer Column Generation , 2005 .

[7]  Bart Selman,et al.  Pushing the Envelope: Planning, Propositional Logic and Stochastic Search , 1996, AAAI/IAAI, Vol. 2.

[8]  Jonathan P. How,et al.  COORDINATION AND CONTROL OF MULTIPLE UAVs , 2002 .

[9]  Gokhan Inalhan,et al.  Large-scale Task/Target Assignment for UAV Fleets Using a Distributed Branch and Price Optimization Scheme , 2008 .

[10]  Maja J. Matarić,et al.  On multi-robot task allocation , 2003 .

[11]  Anthony Stentz,et al.  Traderbots: a new paradigm for robust and efficient multirobot coordination in dynamic environments , 2004 .

[12]  Nidhi Kalra,et al.  Market-Based Multirobot Coordination: A Survey and Analysis , 2006, Proceedings of the IEEE.

[13]  Paul Davidsson,et al.  Agent-Based Dantzig-Wolfe Decomposition , 2009, KES-AMSTA.

[14]  Sebastian Thrun,et al.  Auction Mechanism Design for Multi-Robot Coordination , 2003, NIPS.

[15]  Niklas Sörensson,et al.  Translating Pseudo-Boolean Constraints into SAT , 2006, J. Satisf. Boolean Model. Comput..

[16]  Yoav Shoham,et al.  Combinatorial Auctions , 2005, Encyclopedia of Wireless Networks.

[17]  Martin W. P. Savelsbergh,et al.  A generic view of Dantzig-Wolfe decomposition in mixed integer programming , 2006, Oper. Res. Lett..

[18]  Vincent Conitzer,et al.  Mixed-Integer Programming Methods for Finding Nash Equilibria , 2005, AAAI.

[19]  David C. Parkes,et al.  Iterative combinatorial auctions: achieving economic and computational efficiency , 2001 .

[20]  François Vanderbeck,et al.  On Dantzig-Wolfe Decomposition in Integer Programming and ways to Perform Branching in a Branch-and-Price Algorithm , 2000, Oper. Res..

[21]  Sven Koenig,et al.  Sequential Incremental-Value Auctions , 2010, AAAI.

[22]  John N. Tsitsiklis,et al.  Introduction to linear optimization , 1997, Athena scientific optimization and computation series.

[23]  Warren B. Powell,et al.  A Stochastic Formulation of the Dynamic Assignment Problem, with an Application to Truckload Motor Carriers , 1996, Transp. Sci..

[24]  J.P. How,et al.  Cooperative task assignment of unmanned aerial vehicles in adversarial environments , 2005, Proceedings of the 2005, American Control Conference, 2005..

[25]  Geoffrey J. Gordon,et al.  Distributed Planning in Hierarchical Factored MDPs , 2002, UAI.

[26]  Sven Koenig,et al.  K-swaps: cooperative negotiation for solving task-allocation problems , 2009, IJCAI 2009.

[27]  Dana S. Nau,et al.  On the Use of Integer Programming Models in AI Planning , 1999, IJCAI.