Friends by Design

Friends - by -Design


Main partners of the project initiative:

- Aarhus Universitet: Marco Nørskov & Johanna Seibt (Philosophy)

- Christian-Albrechts-Universität zu Kiel: Oliver Niebuhr (Analysis of Spoken Language) & Christian Kaernbach (Psychology)

- Universität Hamburg: Carola Eschenbach & Christopher Habel (Computer Science)

- Syddansk Universitet: Kerstin Fischer (Linguistics), Norbert Krüger (Robotics), & Klaus Robering (Information Science)



"Friends by Design: Formal Models of Human-Robot Interactions" unites researchers from robotics, computer science, information studies, linguistics, phonetics, psychology, and philosophy for an interdisciplinary study concerned with "sociable" robots. (The project title is borrowed from J.P. Sullins' 2009 article "Friends by Design: A Design Philosophy for Personal Robotics Technology".) Sociable robots are supplied with an amount of "social intelligence" that renders them able to interact, communicate, and relate to their human users in a personal way such that the user might ultimately interact with her/his robot as a companion or friend. The activity is, more precisely, concerned with the development of formal models which can be used both as explanative tools for the analysis and planning of the interaction between sociable agents and as guidelines for the design of sociable robots and their behaviour. The amount of the on-going efforts within academic and industrial research in the field of social robotics clearly indicate that it may well be expected that social robots will gain more and more importance within the next decades especially within the welfare sector.

In order to achieve their goal, social robots must be equipped with a wide spectrum of abilities which roughly fall into three classes: (1) physical, (2) intellectual, and (3) social abilities. "Friends by Design" aims at stimulating research in each of these three classes, in this way continuing research carried out within "Making Space: The Ontology of Social Interaction", which has been funded in 2012 by a "seed money grant".

(1) Physical

  • Social Dimensions of Physical Activities of Robots. If robots perform physical tasks (grasping a cup, lifting a tray, moving through the room) within the space of social human interaction (a living room),the physical activities of robots are peceived as social actions. What are the basic social algorithms that robots should use in various contexts?

  • (2) Intellectual

  • Decision Making and Planning Based on Social Perception. How does the social perception of complex actions derive from that of the combined basic activities? How are actions arranged into comprehensive plans for achieving some given social goal? How can the choice made between different courses of actions base on the evaluation of their social consequences? How should we conceive of "robot actions" from the perspective of moral philosophy?

  • (3) Social

  • Interaction and Communication between Robots and Humans. How do humans interpret signals of robots (e.g., light, sound, gestures, etc) as attempts of social interaction? How do accompanying signals modify the social perception of physical activities?
  • Emotion and Affection and their Role in Human-Robot Interaction. What are the emotional effects of robots upon human beings and how do these effects influence the interaction and communication between humans and robots? Should (and can) the outer appearance of robots be designed in such a way that they trigger (certain) emotions in their users? How can robots detect and identify human emotions? And can, conversely, robots simulate (or even "exhibit") emotions? What are the perspectives of "affective computing"?
  • Applications of Robots in Welfare. What are the special principles (technical, social, and ethical) governing the use of robots in the welfare sector? In what sense can robots "care" for people? To what extend can a robot fill the social role of a caregiver? To what extend can a robot mediate between a human caregiver and the person cared for? What is the exact nature of the relationships between robots, the people served by or cared for by the robots, and institutions or human caregivers delegating care tasks to the robots? What role do affections and emotions play in the constitution of these relationships?

  • Methods

    Social Robotics is a new and increasingly transdisciplinary research area that poses special methodological problems. The research questions listed above are treated with very different methodologies, including conceptual analysis, the construction of suitable formal languages, anthropological observation and field work, psychological and linguistic experimentation as well as technical engineering. One of the main tasks of the project group will be to clarify the methdology of transdisciplinary social robotics, which is a crucial prerequisite for successful grant proposals and more generally a current desideratum in the field.