Demands for robots are growing rapidly recently. However, such demands are wide-ranging, and each demand contains subtle differences. Hence, a rapid and easy robot design method is needed for dealing with a wide variety of demands. A robot design method that fulfills the robot’s inside with a three-dimensional (3D) diamond crystal structure-shaped structure was proposed previously. It is suitable for 3D printing and makes robot designing simple and easy. This proposed method achieved fast and easy robot design, but the structure inside the robot might become weak depending on the appearance of the robot, especially in a tight and thin shape. The method to detect and reinforce such weak structure is proposed in this paper for dealing with this problem using a Cohesive Blocking technique. The effectiveness of the proposed method is evaluated through the experiments of two types of robot appearance models.

In this paper, we developed a reliable set of utterance intention tags and constructed an interview dialogue corpus annotated with these tags. The tagging scheme consists of two hierarchical levels. The first level includes eight broad categories, such as “question,” “answer,” “backchannel,” and “impression.” The second level provides a more fine-grained classification with fourteen tags, including “question-preparation,” “question-opinion,” and “answer-self-disclosure (opinion).” This hierarchical structure was designed to analyze the interviewer’s technique and then to apply these findings into the interview dialogue system. Each dialogue in the corpus was annotated by three to five annotators using the designed tags. It is observed that inter-annotator agreement was evaluated and found to be high, which indicates the strong reliability of tagging on the utterance. Further, we also investigated the cases of annotator disagreement. Specifically, we examined the most common cases where annotators disagreed on tags assigned to interviewer and guest utterances. Our analysis revealed that certain interviewer utterances were deliberately ambiguous, causing difficulty for annotators in assigning definitive tags.

In order for robots and virtual agents to interact naturally with humans, it is essential to build common ground with users. Few previous studies have investigated the process of building common ground in human interaction, and most of them are based on text chat. In this study, we focus on the modality of dialogue and social relationships among speakers and investigate the effects of these factors on the process of building common ground. Specifically, we collected and analyzed dialogues in which common ground was constructed among speakers under the conditions that the modality of the dialogues was audio or video and the social relationship was first meeting or friend. The results showed that the extension of modality and deep social relations accelerate the construction of common ground. Analysis of verbal behaviors also showed that deep social relationships increased the amount of speech required for task progression, and the extension of modality facilitated the transmission of empathic intentions, suggesting that reproducing dialogue acts between friends and nonverbal behaviors that convey empathic intentions are likely to be important when designing robots and virtual agents.