A market-based formulation of sensor-actuator network coordination

That is, the system operates most efficiently when the individuals involved act autonomously based on local information. Given the scale at which such economies operate, the problem of coordinating the actions of all the participants would be intractable without this compartmentalization of data and control. However, it is not the case that there is no communication at all, for without any transfer of information there could be no coordinated activity in the system. Rather the participants in the economy communicate constantly, but through a remarkably efficient medium: price. All information regarding the availability and value of a resource is compressed into a single linear comparator, allowing rational efficient decisions regarding economic transactions. For example, when buying an avocado at the grocery store, we may not know that a drought recently devastated many crops or that avocados are extremely popular right now, but we do know that the store is charging $5 per avocado and we make our buying decision based on that information. We need not take into account the complex (and largely irrelevant) market forces behind the price. As we search for design methodologies that allow coordination in distributed sensor-actuator networks of a scale comparable to human economies, we take inspiration from the price-based market model. The problems central to economics and synthetic network control are extremely similar. In fact, the fundamental issue in both areas is resource allocation: how should the available resources be distributed among the members of the group?

[1]  Lynne E. Parker,et al.  ALLIANCE: an architecture for fault tolerant, cooperative control of heterogeneous mobile robots , 1994, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'94).

[2]  Michael P. Wellman,et al.  The WALRAS Algorithm: A Convergent Distributed Implementation of General Equilibrium Outcomes , 1998 .

[3]  Ross A. Gagliano,et al.  Auction allocation of computing resources , 1995, CACM.

[4]  Katia P. Sycara,et al.  Distributed Intelligent Agents , 1996, IEEE Expert.

[5]  Russell Smith.  The contract net protocol , 1980 .

[6]  Michael P. Wellman,et al.  A market protocol for decentralized task allocation , 1998, Proceedings International Conference on Multi Agent Systems (Cat. No.98EX160).

[7]  Lynne E. Parker,et al.  ALLIANCE: an architecture for fault tolerant, cooperative control of heterogeneous mobile robots , 1994 .

[8]  L. Walras.  Elements of Pure Economics , 1954 .

[9]  Sarit Kraus,et al.  Task Allocation Via Coalition Formation Among Autonomous Agents , 1995, IJCAI.

[10]  Douglas B. Moran,et al.  The Open Agent Architecture: A Framework for Building Distributed Software Systems , 1999, Appl. Artif. Intell..

[11]  Wei-Min Shen,et al.  A Dynamic Distributed Constraint Satisfaction Approach to Resource Allocation , 2001, CP.

[12]  Maja J. Mataric,et al.  Principled Communication for Dynamic Multi-robot Task Allocation , 2000, ISER.

[13]  G. SmithR..  The Contract Net Protocol , 1980 .

[14]  Maja J. Matarić,et al.  Designing emergent behaviors: from local interactions to collective intelligence , 1993 .

[15]  Rahul Simha,et al.  A Microeconomic Approach to Optimal Resource Allocation in Distributed Computer Systems , 1989, IEEE Trans. Computers.

[16]  Milind Tambe,et al.  Agent Architectures for Flexible, Practical Teamwork , 1997, AAAI/IAAI.

[17]  Maja J. Matarić,et al.  Sold!: Market methods for multi-robot control , 2001 .

[18]  Maja J. Mataric,et al.  Pusher-watcher: an approach to fault-tolerant tightly-coupled robot coordination , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).