Model-Based Attention Acquisition Strategies for Socially Interactive Robotics

Copyright is held by the author/owner(s). Workshop on Playing Together, IDC’17, June 27-30, 2017, Stanford, CA, USA. Introduction Roboticists in the field of human-robot interaction (HRI) are seeking to develop algorithms and methods for interactive robots that teach, help, and play with human users [6]. Our work is exploring facets of joint attention, a foundational social skill that is learned in the early stages of development [10] and is of special interest to researchers in fields such as social skill therapy for children with autism [1]. Most human-robot interaction systems use cues such as calling the user by name or making other salient sounds, all of which are known to be effective for attention acquisition [7, 12], a necessary step in joint attention.