Artificial Emotion Generation Based on Personality, Mood, and Emotion for Life-Like Facial Expressions of Robots

摘要

We can't overemphasize the importance of robot's emotional expressions as robots step into human's daily lives. So, the believable and socially acceptable emotional expressions of robots are essential. For such human-like emotional expression, we have proposed an emotion generation model considering personality, mood and history of robot's emotion. The personality module is based on the Big Five Model (OCEAN Model, Five Factor Model); the mood module has one dimension such as good or bad, and the emotion module uses the six basic emotions as defined by Ekman. Unlike most of the previous studies, the proposed emotion generation model was integrated with the Linear Dynamic Affect Expression Model (LDAEM), which is an emotional expression model that can make facial expressions similar to those of humans. So, both the emotional state and expression of robots can be changed dynamically.