This thesis aims to provide a novel framework for a multiagent system implementation. The major feature of the proposed architecture is the introduction of the biological concept of hormones. The hormones are passed via the communication network to convey limited global system state knowledge. The agents' response to a hormone is interpreted depending on its own local agent state. The primary focus of this thesis is the development of the particulars of the architecture. Prior work of multiagent systems research is reviewed and studied for contributions. Biological studies of hormones are employed to draw out interaction rules and analyze control mechanisms in a biological organism. The hormonal communication and control architecture is constructed, with major components detailed by flowcharts. The proposal is tested with two simulations: A minesweeping problem that has been modeled by other models, and an application of the architecture to a hypothetical ant colony. Research on biological ants is presented to suggest the behavior and goals of a model configured to employ the HCC architecture. The model is fleshed out, and the decisions made by considerations to the architecture are explained. The implementation of the simulation programming with the SWARM programming libraries for the Objective-C language is discussed. The data from experimental runs are analyzed with attention to global action.

Library of Congress Subject Headings

Intelligent agents (Computer software); Multiagent systems; Distributed artificial intelligence; Socioendocrinology--Computer simulation; Ants--Behavior--Computer simulation; Minesweepers--Computer simulation

Publication Date


Document Type


Student Type


Degree Name

Electrical Engineering (MS)

Department, Program, or Center

Electrical Engineering (KGCOE)


Vincent Amuso

Advisor/Committee Member

Daniel Phillips

Advisor/Committee Member

Ferat Sahin


Physical copy available from RIT's Wallace Library at QA76.76.I58 T46 2004


RIT – Main Campus