The field of Human-Robot Interaction (HRI) is dedicated to designing, creating, and developing algorithms and methods for evaluating robotic systems to be used by or with human users. Interactions between humans and robots are incredibly complex; researchers in the field are exploring a broad range of problems and approaches, from robot design, to proxemics, to task sharing. Interactive robots ... [Show full abstract] must be able to both perceive and generate rich, engaging, multi-modal social interaction to truly be effective.
We were initially inspired by the needs of children with autism spectrum disorder (ASD), a developmental disorder associated with atypical development of social skills along with other symptoms. One of the social skills that children with ASD struggle with is joint attention, the sharing of focus between individuals during interaction that allows for effective communication and sharing of knowledge and experiences. In our project, we hope to give robots the ability to not only engage in but also explicitly establish joint attention. To do this we first came up with a three-step approach for initiating joint attention and then focus on designing robot behaviors for the first of those three steps, attention acquisition. By generating minimal, yet effective, attention acquisition behaviors, we can reduce the risk of over- whelming users with too much sensory information and better preserve the agency of the robot, a feature for general HRI but especially important for interactions with children with ASD who often have heightened sensitivities to sound and light. Al- though this work was initially targeted towards improving joint attention in children with ASD, joint attention is beneficial in all human-robot interactions and the result- ing approach is general enough to be used across different interaction domains and with many different robot embodiments.