Characterization of Human Trust in Robot through Multimodal Physical and Physiological Biometrics in Human-Robot Partnerships

Abstract

Trust is an attribute that many people use daily, whether consciously thinking of it or not. Although commonly designated as a firm belief in reliability, trust is more complex than many think. It is not just physical, but rather an emotion, feeling, or choice that has many layers, and can be influenced in a variety of ways. As robotics and artificial intelligence grow, humans have to deliberate whether they trust working with these technical counterparts or not. In this work, we build computational models to quantitatively characterize and analyze humans' trust in robots using multimodal physical and physiological biometric data based on the TrustBase we have created through user studies in human-robot collaborative tasks. During human-robot collaborative processes, we have collected physical and physiological attribute data of human subjects as well as the users' trust levels for each interaction. This data is used to develop a database known as TrustBase. With the data from TrustBase, computational and analytical approaches are used to investigate the correlation between robot performance factors and humans' trust levels and to characterize humans' trust in robots during human-robot collaboration. 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.