The role of expectations and situations in human-robot interaction

Abstract

In human-robot interaction, the behaviour of the robot is usually predefined and can easily be predicted. However, this is not true for the behaviour of the users. Throughout the interaction, users employ various behaviours which often cannot be foreseen intuitively. Therefore, theoretical models need to be developed to predict and explain the users’ behaviour in order to enable the robot to better understand its interaction partner. The current work suggests such a model which is based on the concepts of situations and expectations. It claims that both concepts support the prediction of user behaviour in the interaction and demonstrates the usefulness of the model in a case study based on data from human-robot interaction in a home tour situation.