Off-campus UMass Amherst users: To download dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users, please click the view more button below to purchase a copy of this dissertation from Proquest.

(Some titles may also be available free of charge in our Open Access Dissertation Collection, so please check there first.)

A mediation-based approach to cooperative, distributed problem solving

Roger T Mailler, University of Massachusetts Amherst

Abstract

This dissertation focuses on using cooperative mediation to solve distributed problems that are shared between cooperative autonomous agents. Cooperative mediation involves controlled centralization of overlapping subproblems in order to quickly converge on solutions. This technique represents a new paradigm in distributed problem solving that simultaneously exploits the speed of centralized algorithms while using the power of distributed algorithms to identify relevant problem substructures. In this dissertation, we introduce the cooperative mediation paradigm and present three new algorithms used to solve various distributed problems. Amongst these algorithms are asynchronous partial overlay (APO) for solving distributed constraint satisfaction problems, optimal asynchronous partial overlay (OptAPO) for solving distributed constraint optimization problems, and scalable, periodic, anytime mediation ( SPAM) for solving real-time, distributed resource allocation problems. These algorithms have been extensively tested and in the case of APO and OptAPO are shown to be superior to all other current techniques used to solve these problems. ^

Subject Area

Computer science

Recommended Citation

Mailler, Roger T, "A mediation-based approach to cooperative, distributed problem solving" (2004). Doctoral Dissertations Available from Proquest. AAI3136757.
http://scholarworks.umass.edu/dissertations/AAI3136757

Share

COinS