Development of multimodal presentation markup language MPML-HR for humanoid robots and its psychological evaluation

Animated agents that act and speak as attendants to guests on shopping web sites are becoming increasingly popular. Inspired by this development, we propose a new method of presentation using a humanoid robot. Humanoid presentations are effective in a real environment because they can move and look around at the audience similar to a human presenter. We developed a simple script language for multimodal presentations by a humanoid robot called MPML-HR, which is a descendant of the Multimodal Presentation Markup Language (MPML) originally developed for animated agents. MPML-HR allows many non-specialists to easily write multimodal presentations for a humanoid robot. We further evaluated humanoid robots' presentation ability using MPML-HR to find the difference in audience impressions between the humanoid robot and the animated agent. Psychological evaluation was conducted to compare the impressions of a humanoid robot's presentation with an animated agent's presentation. Using the Semantic Differential (SD) method and direct questioning, we measured the difference in audience impressions between an animated agent and a humanoid robot.