AvatARoid: A Motion-Mapped AR Overlay to Bridge the Embodiment Gap Between Robots and Teleoperators in Robot-Mediated Telepresence

Abstract

Robot-mediated telepresence promises to facilitate effective social interaction between remote teleoperators and on-site users. However, disparities between the robot's form and the teleoperator's representation cause perceptual conflict in on-site users, degrading interaction quality. We introduce AvatARoid, a novel design that bridges this embodiment gap by superimposing the teleoperator's motion-mapped AR avatar overlay on a humanoid. We evaluated our design in a mixed-method study (n=48) using an immersive simulation where participants interacted with a confederate teleoperator, presented in either (a) a humanoid robot, (b) a humanoid robot with video, or (c) AvatARoid. Results suggest AvatARoid significantly improved teleoperator embodiment for on-site users, particularly enhancing co-location, and control perceptions, and providing richer non-verbal gestures. In contrast, video and baseline conditions often resulted in a pronounced disconnect between the teleoperator and the robot for on-site users. Our study offers new insights into designing novel teleoperator representations to promote social interaction in robot-mediated telepresence.