GROUP ATTENTION CONTROL FOR COMMUNICATION ROBOTS

Abstract

A humanoid robot can support people in a real environment by interacting with them through human-like body movements, such as shaking hands, greeting, and pointing. In real environments, a robot often interacts with groups of people to provide services, but one-to-many interaction is quite different from one-to-one interaction. For example, a robot cannot satisfy different demands of people simultaneously when it interacts with many people. To solve such problems of interaction, we focused on purpose of attention; it explains why people are aiming their attention in a particular direction. This paper describes a group attention control (GAC) system that enables a communication robot to simultaneously interact with many people. The system is designed to coordinate people's purpose of attention through behavior that is based on two design policies: controlling cooperative situations and indicating explicit control. We implemented a semi-autonomous GAC system in a communication robot that guides visitors to exhibits in a science museum and engages in free-play interactions with them. We developed the semi-autonomous system to concentrate on the behavior-generating mechanism in the GAC system, and to avoid difficulties of sensing problems in real environments. We investigate the effectiveness of the GAC system through a two-week field trial in the museum. Experimental results revealed that visitors evaluated the robot with the GAC system more highly than without it. We believe these results will allow us to develop interactive humanoid robots that can interact effectively with groups of people.