Preliminary Thoughts on Designing and Evaluating Affective Human-Robot Interaction - Part I

Affective robots, which interact with humans in fields like education, healthcare, and companionship, require both cognitive and emotional intelligence. Cognitive intelligence enables robots to perceive and understand human behavior, while emotional intelligence allows robots to empathize with and respond to human emotions. This article outlines the affective capabilities necessary for affective robots, including affective recognition and analysis, emotional strategy, emotional expression, state maintenance, proactive reaction, and safety and ethics, as criteria for evaluating these robots. In the challenges and practice section of this article, we also list selected works from the ACM/IEEE International Conference on Human-Robot Interaction (HRI) that focus on affective interaction, providing a preliminary foundation for these criteria. We will continue to update this article with more studies from recent literature and a broader range of studies in the field. Although this article offers conceptual insights into affective human-robot interaction, it is NOT an academic paper.