Affective Communication Model with Multimodality for Humanoids

Abstract

A humanoid robot that can naturally interact with humans must be capable of high-level interaction skills; that is, it must be able to communicate with humans in the form of humanlike skills such as gestures, dialogue communication with mutual sympathy. In terms of a human symbiotic situation, it is necessary for a robot to be sociable in terms of thinking and feeling. In this aspect, social interaction and communication that share emotions are important issues in the field of a humanoid robot research. We propose an affective communication model for humanoid robots, so that these robots achieve high level communication through the framework of the affective communication model with multimodality. The proposed model comprises five subsystems: namely, perception, motivation, memory, behavior, and expression. In the perception system, we implemented a bimodal emotion recognizer. To ensure a humanoid robot can respond appropriately to the emotional status of users and itself, we designed subsystems that use their own drive, emotions, and memory. The major improvement in our framework is the construction of a memory system that stores explicit emotional memories of past events. The literature from cognitive science and neuroscience suggests that emotional memories are vital when the human brain synthesizes emotions (Ledoux, J., 1996). While previous research on sociable robots either ignores the emotional memory or maintains the emotional memory implicitly in high-level interaction, we need to establish explicit memories of emotions, events, and concepts. Therefore we have adopted the concept of emotional memory for our humanoid robot. Our memory system maintains explicit memories of previous emotional events. Thus, the emotional system can synthesize emotions on the basis of emotional memories. Since 1999, AIM lab directed by Hyun S. yang in KAIST has focused on building new humanoid robots with a self-contained physical body, perception to a degree which allows the robot to be autonomous, and social interaction capabilities of an actual human symbiotic robot. This study was built on previous researches about the social interactions and the developments of the first generation humanoid robots, AMI, AMIET coupled with a humansize biped humanoid robot, AMIO in the AIM Lab (Yong-Ho Seo, Hyun S. Yang et al., 2003, 2004, 2006). The proposed model was used to enhance the interaction between human and humanoid robot. Accordingly, we designed and implemented the methods which are mentioned above