Abstract:
What factors influence people’s preferences for robot assistance during human-robot collaboration tasks? Answering this question can help roboticists formalize definitions of assistance that lead to higher user satisfaction and increased user acceptance of assistive technology. Often in human robot collaboration literature, we see assistance paradigms that aim to optimize task success metrics and/or measures of users’ perceived task complexity and cognitive load. However, frequently in this literature, participants express their preference for paradigms that do not perform optimally with respect to these metrics. Therefore, task success and cognitive load metrics alone do not encapsulate all of the factors that inform the users’ needs and/or desires for robotic assistance. In this work, we aim to study some potential factors that influence users’ preferences for assistance in the domain of assistive teleoperation during object manipulation tasks. By assistive teleoperation (or shared control), we mean a system that assists the user by combining the control signals from the user and the automated assistance computed by the system. In this domain, we study the role that two potential factors (magnitude of end effector movement and the degrees of freedom being controlled) play in influencing the amount of automated assistance the user wants during a teleoperated object manipulation task.

Committee:
Henny Admoni
Zackory Erickson
Oliver Kroemer
Roshni Kaushik