Doctoral Dissertations

Keywords and Phrases

Defect Detection; Generalized Additive Model; Haul Road; Principal Curve; Unmanned Aerial Vehicles; Vibration

Abstract

Mine haul roads degrade rapidly due to extremely heavy loads on sub optimal construction materials. This research tests two methods of defect detection to determine limitations to current technology: optical detection and vehicle vibration response. Optical defect detection was found to have the least significant limitations for surface defect detection and vehicle vibration response modeling has the least significant limitations for subsurface defects. This research proposes novel methods to eliminate the critical limitations of both technologies.

Unmanned Aerial Vehicles (UAV) equipped with GPS and a high-resolution camera is used to create a 3-D point cloud using a structure-from-motion algorithm. This process is effective on paved roads, but not on mine haul roads with poorly defined edges. This research creates a grid system aligned to haul truck GPS traces to define the analysis area. Multiple linear regression produces three metrics for road quality analysis: roughness, wheel path grade, and cross slope grade. This method provides qualitative measurements of road condition verified by visual assessment from the haul truck.

Vibration response is a common detection method due to its low cost and continuous operation. However, dynamic truck movements and operators driving decisions to avoid highly defective road sections limit the effectiveness. This research incorporates turning rate and lateral lane movement in a GAM. The correction factors from this method improve the relationship between measured vibration and road quality especially in road sections with highly dynamic vehicle traffic like corners and intersections”--Abstract, page iii.

Advisor(s)

Galecki, Greg

Committee Member(s)

Awuah-Offei, Kwame, 1975-
Xu, Guang
Liu, Jenny
Hilgers, Michael Gene

Department(s)

Mining Engineering

Degree Name

Ph. D. in Mining Engineering

Publisher

Missouri University of Science and Technology

Publication Date

Fall 2021

Pagination

xii, 85 pages

Note about bibliography

Includes bibliographic references (pages 75-84).

Rights

© 2021 Alexander David Douglas, All rights reserved.

Document Type

Dissertation - Open Access

File Type

text

Language

English

Thesis Number

T 11944

Download

Share

 
COinS