[4C3-GS-11] AI and Society:

[4C3-GS-11-05] Robot System that can Give Route Directions with Landmarks

〇Kanta Tachikawa¹, Shota Akahori¹, Yuita Sakuma¹, Michita Imai² (1. Keio University Graduate School, 2. Keio University)

Keywords:HRI, NLP

In the task of providing information to a person, it is important to make the person understand the information, and in the task of giving route directions, it is necessary for the robot to use utterances and gestures to give the route. However, most previous works related to giving route directions robots only assume that robots can give route directions from the current location, and robots give the complicated route even if the route from the current location is complex. In this paper, we propose a robot system, RoDiL (Route Directions with Landmarks), which gives a simple route using landmarks, based on the user's city knowledge such as the location of stores and buildings, if the route from the current location to the destination is complex. 100 people evaluated "directions with landmarks based on RoDiL" compared to "directions without landmarks", and the results showed that when the route from the current location was complex, the directions with landmarks were rated significantly higher than without landmarks. When the route from the current location was simple, the directions without landmarks gave a better impression, indicating that RoDiL's design criteria were highly evaluated.