Synchronized multimodal expression generation using editing toolkit for a human-friendly robot

摘要

Attempts to put robots to practical use have been increased, as robots become more human-friendly. In the human-robot interaction field, main issues are how variously the robot can express its emotion and how much the expression is socially acceptable. This paper proposed the editing toolkit which allows us to simulate a 3D model robot in order to express robot's emotions and intentions for human-robot interaction and robot services. Using the editing toolkit, we have generated multimodal expressions and formulated the method to combine a few of the primitive expressions. The robot, which we used for simulation, has three modalities such as a facial expression, a neck motion and a gesture with two arms. The expressions of each modality were used for generating multimodal expressions, synchronizing with the time information obtained from a professional actor. Consequently, for three emotions and thirteen intentions of the robot, we have generated primitive expression database set and synchronized multimodal expression using editing toolkit.