Date of Award

5-2026

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

Debapriya Hazra

Committee Member

Junyi Ye

Abstract

Robot systems exist in prevalence across numerous industries, such as automotive, agriculture, and manufacturing. In many cases, these systems are fully autonomous, being products of the Industry 4.0 rush that produced a surge in robotics. Now, as the world begins to transition towards the human-centric Industry 5.0, collaborative systems involving both human and machine are becoming more commonplace. Such collaborative systems generally consist of humans and robots sharing both tasks and workspaces to accomplish a common goal. As expected, it is important for human workers to find their robot partners trustworthy and acceptable and to be comfortable working with them. Understanding the factors that influence each of these facets, as well as their interdependence, can facilitate more optimal human-robot collaborative configurations. This focus becomes even more relevant when considering Multi-Human Multi-Robot contexts which, by nature, are more complex working environments. In this work, the three facets of trust, comfort, and acceptance are investigated according to various factors, including contextual (human-robot configuration), individual (factors relevant to one’s experiences and perceptions), and physiological, through a multifactored user study. Predictive capabilities, effect directions and magnitudes for each are modeled according to a hierarchical Bayesian ordinal regression framework in order to uncover associations between them. Results suggest that both trust and comfort are most significantly associated with variations in individual factors such as attitude towards robots and acceptance of behavior, while the configuration of the collaborative setting itself is more moderately associated. Physiological features expressed relatively smaller effect sizes yet still provide valuable insight into ideal features to track when observing real-time trust and comfort levels in human-robot collaboration.

File Format

PDF

Available for download on Saturday, June 03, 2028

Share

COinS