Emotion-sensitive robots - a new paradigm for human-robot interaction

Abstract

An emotion-sensitive human-robot cooperation framework where a robot is sensitive to the emotions of the human working with it and is also capable of changing its behavior based on this perception is presented in this paper. Peripheral physiological responses of a human are measured through wearable biofeedback sensors to detect and identify his/her underlying level of anxiety. A control architecture inspired by Riley's original information-flow model is designed. In this human-robot interaction framework, the robot is responsive to the psychological states of the human and detects both implicit and explicit communication from the human to determine its own behavior. Human-robot cooperation experiments using a mobile robot as a test bed are performed where the robot senses anxiety level of the human and responds appropriately. The results presented here validate the proposed framework and demonstrated a new way of achieving emotion-based interaction between a human and a robot.