Infant-like social interactions between a robot and a human caretaker. Special Issue of Adaptive Behavior on Simulation Models of Social Agents

This paper presents an autonomous robot designed to interact socially with human “par-ents”. A human infant’s emotions and drives play an important role in generating meaningful interactions with the caretaker, regulating these interactions to maintain an environment suit-able for the learning process, and assisting the caretaker in satisfying the infant’s drives. For our purposes, the ability to regulate how intensely the caretaker engages the robot is vital to successful learning in a social context. To achieve a similar interaction dynamic, we present a general framework that integrates perception, attention, drives, emotions, behavior selection, and motor acts. We then present a specific implementation of this architecture which enables the robot to perceive both salient so-cial stimuli (faces) and salient non-social stimuli (motion). The robot responds with expressive displays which reflect an ever-changing motivational state and which give the human cues on how to satisfy the robot’s drives while neither over-whelming nor under-stimulating the robot. Results from a series of experiments are presented where a human engages the robot in either direct face-to-face exchanges or with a toy. We believe this work is an important step toward realizing autonomous robots that can engage in meaningful bi-directional social interactions with humans