Effect of emotional synchronization using facial expression recognition in human-robot communication

In this paper, we developed a KANSEI communication system between human and robot based on emotional synchronization to human emotional state using facial expression recognition. And we conducted experiments to evaluate the effectiveness of the proposed system. The robot recognizes human emotion through their facial expressions, and synchronizes its own emotion with the recognized emotion dynamically by using a vector field of dynamics. Then the robot expresses its own emotions by facial expressions. In the communication experiment, we found that the subjects´ feeling became much comfortable after communicating with the robot in the case of emotional synchronization, while that the subjects felt a little uncomfortable after communicating with the robot in the case of non-synchronization. Meanwhile, subjects in the case of emotional synchronization communicated much more with the robot, and the communication time was as double times as the time when communicated in the case of non-synchronization. Furthermore, subjects communicated in the case of emotional synchronization had good impressions on the robot, and also were much better than the impressions in the case of non-synchronization. So it was confirmed in this study that the emotional synchronization in human-robot communication could be effective to make human keeping a comfortable state and would make the robot much more favorable and acceptable to human.