Aarhus University Seal / Aarhus Universitets segl

On the pertinence of Social Practices for Social Robotics

Aurélie Clodic, LAAS-CNRS, Université de Toulouse, CNRS, Toulouse, France
Rachid Alami, LAAS-CNRS, Université de Toulouse, CNRS, Toulouse, France
Virginia Dignum, Delft University of Technology, Netherlands
Agnese Augello, Istituto di Calcolo e Reti ad Alte Prestazioni, National Research Council of Italy, Italy
Frank Dignum, Utrecht University, Utrecht, The Netherlands
Javier Vázquez-Salceda, Universitat Politècnica de Catalunya (UPC), Spain
Manuel Gentile, Istituto per le Tecnologie Didattiche, National Research Council of Italy, Italy

In the area of consumer robots that need to have rich social interactions with humans, one of the challenges is the complexity of computing the appropriate interactions in a cognitive, social and physical context.

Of course people have managed this complexity for ages already. One of the ways to simplify social interactions is by standardizing them based on particular contexts. E.g. even though greetings have many variations, the patterns they follow are quite standard and used in some form all over the world. At the same time the variations within the physical actions that can be chosen have their own social effect. E.g. boxing a hand instead of shaking it signifies that we are in an informal setting and are peers in this context.

We chose the human use of social practices and its associated theory as a basis for modeling the interactions for social robots. Social Practices describe physical and social patterns of joint action as routinely performed in society and provide expectations about the course of events and the roles that are played in the practice.

We will show how Social Practices can structure human-robot interactions in a way that feels natural to people and can ensure that robots are aware of their own social identities and the identities of others and also be able to identify the different social contexts and the appropriate social practices in those contexts.