The affective loop: A tool for autonomous and adaptive emotional human-robot interaction

Abstract

The paper presents an affective model for social robotics, where the robot is capable of behavior adaptation, in accordance with the needs and preferences of a particular user. The proposed approach differs from other studies in human-robot interaction as these usually have been using the ‘Wizard of Oz’ technique, where a person remotely operates a robot. On the other side, simulated robots are not able of personalized behaviors and behave according to the preprogrammed set of rules. We provide a tool to personalize affective artificial behaviors in cooperative human—robot scenarios, where human emotion recognition, appropriate robotic behavior selection and expression of robotic emotions play a key role. The preliminary experiments show that the personalized affective robotic behavior can achieve better results in a scenario in which a robot motivates children in learning. We believe that human—robot interfaces which mimic how humans interact with one another in an empathic way could ultimately lead to robots being accepted in the wider domain.