Date of Award

5-2024

Document Type

Thesis

Degree Name

Master of Science (MS)

College/School

College of Science and Mathematics

Department/Program

School of Computing

Thesis Sponsor/Dissertation Chair/Project Chair

Weitian Wang

Committee Member

Hao Liu

Committee Member

Rui Li

Abstract

With recent advancements in machine learning and artificial intelligence, we have entered a new era of technology. With AI on the rise, it has led to many insurgencies in a variety of domains. One area that has particularly been growing at a rapid pace is robotics. Consequently, the ubiquitous rise of robots in new workspaces leads to the influx of human-robot collaboration. Human-robot collaboration is an important factor in this domain, as it mandates the relationship between humans and robots. One key component of this relationship is trust, which many people subconsciously determine when interacting with peers, strangers, and coworkers. Although commonly designated as a firm belief in reliability, trust is rather more complex and can be thought of to have layers that each hold different meanings. Trust can be more than just a cognition, trust can be physical or mental; it can be an emotion, feeling, or choice, depending on the environment. This brings forth the motivation for this work, as we developed a database known as TrustBase, using multimodal sensors to acquire physical and physiological biometric information. These sensors collected data such as electrocardiography, electromyography, optics, and electroencephalography. With the data from TrustBase, computational and analytical approaches, including TabPFN, XGBoost, and Support Vector Machines, are used to investigate if these features have any correlation to trust, and using this data, we can classify and predict whether a user will trust their robot counterpart. Additionally, we bring forth a novel mathematical model to do additional analysis, and ultimately compare to the standard aforementioned machine learning techniques. Results and their analysis suggest the effectiveness of the developed models, providing new findings to the human factors and cognitive ergonomics in human-robot interaction. Future research directions are also discussed.

File Format

PDF

Available for download on Wednesday, November 04, 2026

Share

COinS