Robotic System Controlling Target Human’s Attention

Abstract

Attention control can be defined as shifting people's attention from their existing direction toward a goal direction. If a human would like to shift another's attention, s/he may first turn his/her gaze to that human to make eye contact. However, it is not an easy task for a robot when the human is not looking at it initially. In this paper, we propose a model of attention control with four parts: attention attraction, eye contact, attention avoidance, and attention shift. To initiate attention control process, the robot first tries to gain the target person's attention toward it through head turning or head shaking action depending on the three viewing situations where the robot is captured in his/her field of view (central field of view, near peripheral field of view, and far peripheral field of view). After gaining his/her attention, the robot makes eye contact through showing gaze awareness by blinking its eyes, and directs his/her attention to an object by turning both its eyes and head. If a person other than the target seems to be attracted by the robot, the robot turns its head away from that person to avoid his/her attention. Evaluation experiments confirmed that the proposed approach is effective to control the target person's attention.