Nonverbal Communication of In-Vehicle Humanoid Robot to Communicate Autonomous Vehicle Behavior

Abstract

Autonomous vehicles have been researched and developed all over the world. However, people are still concerned about the safety and comfort of an autonomous vehicle. One of the reasons is that a passenger is difficult to predict the behavior of an autonomous vehicle. In this study, we propose to use a small size humanoid robot as an in-vehicle interface to bridge the communication between a passenger and an autonomous vehicle. An in-vehicle humanoid robot that adopts driving behavior like a human driver transmits autonomous vehicle behavior information by encouraging the prediction of vehicle behavior to a passenger. This paper shows how a robot's motion affects a passenger's prediction of autonomous vehicle behavior. Through the experiments, we discuss the relationship between the motion timing of a robot and a passenger's perception of vehicle behavior. Moreover, we show that a robot motion leads gaze direction to attention object direction based on gaze direction analysis.