Dissertation/Thesis Abstract

Robot configuration for subterranean modeling
by Omohundro, Zachary Meyer, Ph.D., Carnegie Mellon University, 2009, 164; 3386044
Abstract (Summary)

Void models are invaluable for understanding subterranean conditions and guiding underground operations, but many maps are inaccurate, missing or difficult to acquire. Underground voids are inaccessible and often hostile to man and machine. Mapping and modeling of subterranean voids is a superb motivation for development of robots. Successful robot configurations for modeling subterranean worlds such as sewers, tunnels, mines and voids have yet to be created and characterized.

This research identifies subterranean modeling tasks and exploration constraints. Modeling tasks range from limited two-dimensional cross sections to comprehensive, globally-referenced, three-dimensional coverage of vast voids. Entry constraints such as small diameter holes limit the size of robot configurations, but once inside, muck and obstacles require significant terrainability that only comes with larger size. Based on these conditions and constraints, robot configurations are conceived, developed, tested and classified.

This research distinguishes three classes of robotic configuration for void modeling based on entry constraint and robot mobility. Portal Explorers are mobile robots featuring substantial locomotion, sophisticated sensing, and plentiful energy storage that can roam afar after being deployed via large diameter portals. Borehole Sensors provide exceptional modeling capability from fixed vantage points, while their miniaturization enables deployment down smallest diameter boreholes. Borehole Explorers combine some mobility with some borehole access miniaturization to acquire otherwise unobtainable model data. Mobility provides greater modeling coverage than possible with borehole sensors, while reconfigurability allows deployment via small diameter boreholes, enabling void access not possible with portal explorers.

This research classifies subterranean modeling tasks, develops robots to address these tasks, and tests these robots by entering and modeling vast voids. Model data is used to evaluate the merits and limits of these important robot configurations. Additionally, field work provides insight into issues and operational approaches that are not otherwise apparent from lab work alone. Finally, future robot configurations and paradigms for subterranean robotics are envisioned.

Indexing (document details)
Advisor: Whittaker, William Red
Commitee:
School: Carnegie Mellon University
School Location: United States -- Pennsylvania
Source: DAI-B 70/12, Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Mining engineering, Robotics
Keywords: Robot configuration, Robot mobility, Subterranean, Void models
Publication Number: 3386044
ISBN: 978-1-109-51343-1
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest