Autonomous construction agents: An investigative framework for large sensor network self-management

Thumbnail Image
Date
2008-01-01
Authors
Koch, Joshua
Major Professor
Advisor
Kenneth M. Bryden
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Organizational Unit
Journal Issue
Is Version Of
relationships.hasVersion
Series
Department
Mechanical Engineering
Abstract

Recent technological advances have made it cost effective to utilize massive, heterogeneous sensor networks. To gain appreciable value from these informational systems, there must be a control scheme that coordinates information flow to produce meaningful results. This paper will focus on tools developed to manage the coordination of autonomous construction agents using stigmergy, in which a set of basic low-level rules are implemented through various environmental cues. Using VE-Suite, an open-source virtual engineering software package, an interactive environment is created to explore various informational configurations for the construction problem. A simple test case is developed within the framework, and construction times are analyzed for possible functional relationships pertaining to performance of a particular set of parameters and a given control process. Initial experiments for the test case show sensor saturation occurs relatively quickly with 5-7 sensors, and construction time is generally independent of sensor range except for small numbers of sensors. Further experiments using this framework are needed to define other aspects of sensor performance. These trends can then be used to help decide what kinds of sensing capabilities are required to simultaneously achieve the most cost-effective solution and provide the required value of information when applied to the development of real world sensor applications.

Comments
Description
Keywords
Citation
Source
Subject Categories
Copyright
Tue Jan 01 00:00:00 UTC 2008