Multiagent expedition with tightly and loosely coupled decision paradigms

dc.contributor.advisorXiang, Y.
dc.contributor.authorHanshar, Franklin Thomas
dc.date.accessioned2020-12-03T18:08:48Z
dc.date.available2020-12-03T18:08:48Z
dc.date.copyright2007
dc.degree.departmentDepartment of Computing and Information Scienceen_US
dc.degree.grantorUniversity of Guelphen_US
dc.degree.nameMaster of Scienceen_US
dc.description.abstractDEC-POMDPs provide formal models of many cooperative multiagent problems, but their complexity is nondeterministic exponential-time complete in general. We introduce a sub-class of DEC-POMDPs termed 'multiagent expedition ' (MAE). A typical instance consists of an area populated by mobile agents which have no prior knowledge of the area, limited sensing ability, and the outcomes of their actions is uncertain. Success in MAE relies on planning actions that result in high accumulated rewards. We solve an instance of MAE based on collaborative design networks, a 'tightly-coupled' decision-theoretic multiagent graphical model, to demonstrate its generality. We compare our approach to a 'loosely-coupled' decision making paradigm, the recursive modeling method (RMM) and greedy agents in terms of solution quality, scalability and knowledge representation. Experimental results demonstrate significant superior performance of our system in comparison to RMM and greedy agents, and provides insights into drawbacks associated with loosely-coupled systems when dealing with problems such as MAE.en_US
dc.identifier.urihttps://hdl.handle.net/10214/21857
dc.language.isoen
dc.publisherUniversity of Guelphen_US
dc.rights.licenseAll items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.
dc.subjectDEC-POMDPsen_US
dc.subjectmodelen_US
dc.subjectmultiagent expeditionen_US
dc.subjectcollaborative design networken_US
dc.subjecttightly-coupled decision-theoretic multiagent graphical modelen_US
dc.subjectloosely-coupled decision making paradigmen_US
dc.subjectrecursive modeling methoden_US
dc.subjectgreedy agentsen_US
dc.subjectsolution qualityen_US
dc.subjectscalabilityen_US
dc.subjectknowledge representationen_US
dc.titleMultiagent expedition with tightly and loosely coupled decision paradigmsen_US
dc.typeThesisen_US

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Hanshar_FranklinT_MSc.pdf
Size:
4.22 MB
Format:
Adobe Portable Document Format