Author ORCID Identifier:
Date of Graduation
12-2025
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Computer Science (PhD)
Degree Level
Graduate
Department
Computer Science & Computer Engineering
Advisor/Mentor
Rainwater, Chase
Committee Member
Cothren, Jackson
Second Committee Member
Gauch, John
Third Committee Member
Le, Ngan
Keywords
Deep Learning; GPS Denied; Localization; Reamining Useful Lifetime; Reliability; Xception
Abstract
In this dissertation, we explore the potential of machine learning and deep learning techniques to enhance the performance and robustness of applications across two major domains. By addressing the challenges within these fields, we demonstrate that we can leverage learning algorithms to obtain substantial improvements in accuracy and robustness. First, we tackle a problem in the field of predictive health maintenance. We propose a novel auto encoder and neural network based methodology to predict failure times in complex aviation systems to learn to distinguish between normal and abnormal operational behavior, and use this information to inform the neural network to make better predictions regarding failure times. Second, we turn to address the challenges posed in remote sensing applications, specifically, absolute and relative visual localization in GNSS-denied areas. First, we introduce a novel methodology for absolute visual localization in high-altitude unorthorectified satellite imagery. This work is the first in the literature to address this domain, and our work achieves state-of-the art results in this application scenario. Finally, the last chapter pivots to address relative visual localization at high altitudes using unorthorectified satellite imagery in GNSS-denied environments. We introduce a novel method that leverages inertial measurement units sensor data and absolute visual localization predictions from a convolutional neural network (Xception). We further present an improved training procedure to reduce outlier predictions in absolute visual localization. By integrating refined Xception predictions with sensor data and a Kalman Filter, our relative localization approach produces smooth, drift-corrected flight path estimations that remain robust even under challenging conditions and noisy satellite imagery.
Citation
Toche Pizano, R. (2025). Artificial Intelligence for Reliability: Predictive Health Maintenance and Geolocation in GPS-Denied Environments. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/5996
