Quantifying the Subjective Experience in Human–Robot Interaction

Abstract

Assessing the quality of interaction between collaborative robots (i.e., cobots) and humans to achieve a (shared) goal is an essential part of supporting the diffusion of usable and safe collaborative robotic systems. The key issue here is that cobots within and between domains can vary substantially in terms of features even when these are meant to support similar tasks. Under this perspective, it seems very hard, if not impossible, to identify coherent ways to benchmark the quality of cobots. However, all the cobots share one element which is that they have to deal with humans and satisfy them. This chapter claims that focusing on how a system can satisfy the human counterpart can (indirectly) provide insights into the quality of cobots that can be a piece of valuable information for the designers. In this sense, identifying and validating a set of key aspects to assess the subjective experience in interaction with robots during collaborative tasks is a stepping stone for benchmarking cobots. We will discuss how, in other domains, researchers developed standardized scales to ensure comparable results regarding the user experience (UX), and to model the differences between systems. Moreover, based on a recent systematic literature review, an initial list of key aspects to evaluate UX with cobots is discussed as a step to build and validate a future tool for unifying how experts evaluate cobots.