The aim of the research team in this scenario is to development a suitable evaluation methodologies to assess – qualitatively and quantitatively – the interaction between humans and different types of autonomous systems moving in shared physical spaces, with a particular focus on mutual action and intention recognition, and to provide a roadmap for future development based on these results.

The main contribution of the framwork for AIR itself lies within the evaluation methodologies. The theoretical framework developed will have an impact beyond the duration of AIR as it fills an important gap and can be expected to play a pivotal role in the future development of the field of research on human interaction with autonomous systems. The evaluation methodologies are inspired from areas including cognitive science, psychology, human-computer interaction, rather than traditional engineering approaches since the focus has to be on the quality of interaction between people and autonomous systems. Areas addressed are:

Theory of AIR

Briefly stated, the purpose of this task is to provide the theoretical basis on which mutual action and intention recognition between humans and different types of autonomous systems can be characterized and understood.

Characterization of common factors in the application scenarios

The purpose of this task is to scrutinize the interactions designed and studied in the preceding work packages to identify common factors underlying AIR across very different application domains and in human interaction with very different types of autonomous agents.

Understanding and evaluating AIR

Mutual action and intention recognition in the interaction between humans and autonomous machines can be evaluated in a number of different ways, the primary examples of which are qualitative (e.g. how is the interaction experienced?) and quantitative.

 

AIR is a Swedish distributed research environment financed by the