How Should a Robot do Indirect Handovers Politely?

Abstract

Abstract In this work, we studied the action of a robot handing over an object to a human. We focused on indirect handovers, where the robot does not pass the object directly into the human’s hand, but instead puts it down somewhere in the person’s vicinity where they can easily pick it up. In particular, we were interested in understanding what makes an indirect handover perceived as polite. Based on an in-lab collection of data from experienced waiters during customer serving, including follow-up interviews, and labelling of characteristics of their service actions, we proposed a basic model of how polite indirect handovers should be performed. We compared the proposed model with handovers using different arm or gaze motions. In both an in-person experiment and a large scale online video survey, based on participant ratings collected using the Robotic Social Attributes Scale (RoSAS), we confirmed that the proposed polite indirect handover model was evaluated as being more warm, while not significantly affecting the perceived competence or discomfort.