Synchronization of decentralized multiple-model systems by market-based optimization

Market-based optimization is a new optimization method for large decentralized systems where the distributed resource allocation of an economic system is adopted. Market-based algorithms can be interpreted as multi-agent scenarios where producer and consumer agents both compete and cooperate on a market of specified commodities. The market-based approach is applied to the synchronization of a set of local multiple-model systems. The method is extended to the case where each of the subsystems is represented by a Takagi-Sugeno (TS) fuzzy system. Although all local systems are provided with the same control input, the behaviors of the local systems are, in general, different because of different parameters in the subsystems. The task of the market-based optimization is to find an appropriate composition of subsystems so that all local systems exhibit a similar dynamical behavior. Examples show that even systems with potentially unstable local systems can be synchronized if there exists a stable combination of weighted subsystems.

[1]  S. Clearwater.  Market-based control: a paradigm for distributed resource allocation , 1996 .

[2]  Peter Liu,et al.  Semi-decentralized adaptive fuzzy control for cooperative multirobot systems with Hinfin motion/internal force tracking performance , 2002, IEEE Trans. Syst. Man Cybern. Part B.

[3]  E. Haddad.  Real-time optimization of distributed load balancing , 1994, Second Workshop on Parallel and Distributed Real-Time Systems.

[4]  Holger Voos,et al.  A new approach for optimal control using market-based algorithms , 1999, 1999 European Control Conference (ECC).

[5]  Christian Gerber,et al.  On the suitability of market-based mechanisms for telematics applications , 1999, AGENTS '99.

[6]  Marek B. Zaremba,et al.  Distributed flow control design for repetitive manufacturing processes , 1999 .

[7]  Ruzena Bajcsy,et al.  Scaling the Dynamic Approach to Path Planning and Control: Competition among Behavioral Constraints , 1999, Int. J. Robotics Res..

[8]  Tad Hogg,et al.  Controls for unstable structures , 1997, Smart Structures.

[9]  Roderick Murray-Smith,et al.  Multiple Model Approaches to Modelling and Control , 1997 .

[10]  Umit Ozguner,et al.  Combined decentralized multi-destination dynamic routing and real-time traffic light control for congested traffic networks , 1999, Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304).

[11]  Dr. Rainer Palm,et al.  Model Based Fuzzy Control , 1997, Springer Berlin Heidelberg.

[12]  Jacqueline Ayel.  Supervising conflicts in production management , 1995 .

[13]  Philippe Lacomme,et al.  Multi-Agents Approach for Manufacturing Systems Optimization , 1996, PAAM.

[14]  James K. Archibald,et al.  A utility approach to multi-agent coordination , 2000, Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065).

[15]  E. Altman,et al.  A team-theoretic approach to congestion control , 1999 .

[16]  V. C. Ramesh,et al.  A multi-agent mixed initiative system for real-time scheduling , 1998, SMC'98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218).

[17]  Steven E. Rhoads.  The Economist's View of the World: Governments, Markets, and Public Policy , 1987 .

[18]  Stéphane Lafortune,et al.  Some issues concerning decentralized supervisory control with communication , 1999, Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304).