Title
Reproducibility and Replicability in Unmanned Aircraft Systems and Geographic Information Science
Date of Graduation
5-2023
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Geosciences (PhD)
Degree Level
Graduate
Department
Geosciences
Advisor/Mentor
Jason A. Tullis
Committee Member
Kátia Fernandes
Second Committee Member
Xiao Huang
Third Committee Member
Edward Holland
Keywords
AI, interactive text bots
Abstract
Multiple scientific disciplines face a so-called crisis of reproducibility and replicability (R&R) in which the validity of methodologies is questioned due to an inability to confirm experimental results. Trust in information technology (IT)-intensive workflows within geographic information science (GIScience), remote sensing, and photogrammetry depends on solutions to R&R challenges affecting multiple computationally driven disciplines. To date, there have only been very limited efforts to overcome R&R-related issues in remote sensing workflows in general, let alone those tied to disruptive technologies such as unmanned aircraft systems (UAS) and machine learning (ML). To accelerate an understanding of this crisis, a review was conducted to identify the issues preventing R&R in GIScience. Key barriers included: (1) awareness of time and resource requirements, (2) accessibility of provenance, metadata, and version control, (3) conceptualization of geographic problems, and (4) geographic variability between study areas. As a case study, a replication of a GIScience workflow utilizing Yolov3 algorithms to identify objects in UAS imagery was attempted. Despite the ability to access source data and workflow steps, it was discovered that the lack of accessibility to provenance and metadata of each small step of the work prohibited the ability to successfully replicate the work. Finally, a novel method for provenance generation was proposed to address these issues. It was found that artificial intelligence (AI) could be used to quickly create robust provenance records for workflows that do not exceed time and resource constraints and provide the information needed to replicate work. Such information can bolster trust in scientific results and provide access to cutting edge technology that can improve everyday life.
Citation
Howe, C. (2023). Reproducibility and Replicability in Unmanned Aircraft Systems and Geographic Information Science. Graduate Theses and Dissertations Retrieved from https://scholarworks.uark.edu/etd/5090