Spatial Scaffolding for Sociable Robot Learning

Abstract

Spatial scaffolding is a naturally occurring human teaching behavior, in which teachers use their bodies to spatially struc-ture the learning environment to direct the attention of the learner. Robotic systems can take advantage of simple, highly reliable spatial scaffolding cues to learn from human teach-ers. We present an integrated robotic architecture that com-bines social attention and machine learning components to learn tasks effectively from natural spatial scaffolding inter-actions with human teachers. We evaluate the performance of this architecture in comparison to human learning data drawn from a novel study of the use of embodied cues in human task learning and teaching behavior. This evaluation provides quantitative evidence for the utility of spatial scaffolding to learning systems. In addition, this evaluation supported the construction of a novel, interactive demonstration of a hu-manoid robot taking advantage of spatial scaffolding cues to learn from natural human teaching behavior.