偶発的な人とロボットの遭遇に影響を与える：表現豊かな動きが四足歩行サービスロボットに対する歩行者の印象を改善する
Influencing Incidental Human-Robot Encounters: Expressive movement improves pedestrians' impressions of a quadruped service robot
実世界で四足歩行サービスロボットに犬らしいボディランゲージを追加し、歩行者222人を対象に印象を比較したところ、全項目で好意的な評価が得られた。
著者: Elliott Hauser, Yao-Cheng Chan, Ruchi Bhalani, Alekhya Kuchimanchi, Hanaa Siddiqui, Justin Hart
分類: cs.RO, cs.HC
原文アブストラクト
A single mobile service robot may generate hundreds of encounters with pedestrians, yet there is little published data on the factors influencing these incidental human-robot encounters. We report the results of a between-subjects experiment (n=222) testing the impact of robot body language, defined as non-functional modifications to robot movement, upon incidental pedestrian encounters with a quadruped service robot in a real-world setting. We find that canine-inspired body language had a positive influence on participants' perceptions of the robot compared to the robot's stock movement. This effect was visible across all questions of a questionnaire on the perceptions of robots (Godspeed). We argue that body language is a promising and practical design space for improving pedestrian encounters with service robots.
