Towards methodological principles for user studies in Human-Robot Interaction

Kathleen Belhassein (1, 2) , Guilhem Buisan (2) , Aurélie Clodic (3) , Rachid Alami (2)

1 CLLE-LTC - Cognition, Langues, Langage, Ergonomie
2 LAAS-RIS - Équipe Robotique et InteractionS
3 LAAS-IDEA - Service Informatique : Développement, Exploitation et Assistance

Robots are becoming more and more present in our everyday life at work or for entertainment. This raised the need for a principled and well-understood interaction between humans and robots which led to the creation of the Human-Robot Interaction (HRI) research field. In order to assess the models, designs and algorithms elaborated in these researches, evaluation techniques have to be employed in the context of user studies. In this work, we aim to adapt from other research fields some principles that might be used in HRI user studies. More precisely we discuss around some frequent issues concerning recruited users, evaluation methods and replication of the studies, and how some methodological practices could circumvent them. We will finally discuss more generally on HRI studies, claiming that they need methods and assessment techniques specific to their particularities.

Mots clés: human-robot interaction; evaluation methods; user studies

Domaines: Robotique [cs.RO]

Kathleen Belhassein, Guilhem Buisan, Aurélie Clodic, Rachid Alami. Towards methodological principles for user studies in Human-Robot Interaction. Test Methods and Metrics for Effective HRI in Collaborative Human-Robot Teams Workshop, ACM/IEEE International Conference on Human-Robot Interaction, Mar 2019, Daegu, South Korea. ⟨hal-02282600⟩