Emotion-based Architecture for Social Interactive Robots

Abstract

Applications of robots as assistance systems in household, health-or elderly care lead to new challenges for roboticists. These robots need to fulfill complex tasks and navigate in complex human "real world" environments. One of the most difficult challenges in developing assistance robots for everyday life is the realization of an adequate human-robot interface. Traditional interfaces like mouse, keyboard, or joystick appear inappropriate because even untrained non-robot or computer specialists should be able to control the robot. From our point of view, the best way to achieve this goal is, to enable the robot to "natural" human interaction. Since approx. 65% of the information in human-human interaction is conducted non-verbally