Cognitive and Social Design of Robotic Assistants

Speaker
Sara Kiesler and Jennifer Goetz
Human-Computer Interaction Institute, Carnegie Mellon University

When
-

Where
Newell-Simon Hall 1305 (Michael Mauldin Auditorium)

Description

In the future, autonomous, mobile robots will assist people in many environments. Robots will assist caretakers of disabled and elderly people, and they will be used in institutional settings such as schools and hospitals to carry out tasks such as serving food or teaching calisthenics. In each of these situations, humans are integral to the system-as remote operators, as client-users, and as others who live and work in the social settings where robots are employed. As the “HCI” component of the Nursebot project, the goals of our research are to contribute to a basic understanding of people’s interactions with interactive robots, to facilitate useful and graceful interactions between people and autonomous robotic assistants, to aid in robotic technology development, and to advance dialogue on some of the social issues surrounding deployment of life-like robots in work and home settings. In this talk, we will describe one piece of this work—our early research on people’s interactions with human-like and not-so-human -like robots. We will outline a preliminary model of how (and why) people form anthropomorphic impressions of interactive robots, and discuss some implications of this work for the design of assistive social robots.

Speaker's Bio

Sara Kiesler is Professor of Human-Computer Interaction in the HCII at Carnegie Mellon University. Her previous work on communication in computer-mediated contexts and on people’s interactions with human-like computer agents kindled her interest in social robots; this interest was greatly encouraged by the engineering and design work on interactive robots at CMU.

Jennifer Goetz is a Research Associate in the Human-Computer Interaction Institute. She is leading the HCI work on the Nursebot project, and has participated in all the initial user tests and contextual inquiry work on that project.