Be FAIR to Pedestrian Dynamics Data

Abstract

For improving the safety of people in large crowds, it is of great importance to understand the basic mechanisms of pedestrian dynamics, e.g. with help of experiments. The number of openly shared datasets of these experiments has increased in the last years also due to stricter requirements from journals and funders. We share our own experimental data by an open access data archive which data is widely used in the community. However, our data and also data of other researchers in the field of pedestrian dynamics is not annotated in a systematic or semantically harmonized way, which impairs FAIRness in general and interoperability specifically. In this paper, we propose a standardized extensible metadata schema and key data structures for trajectories and geometry. The proposed metadata schema and data structures hopefully support the interoperability within the community and will assist to make data reutilization more efficient. Our own legacy datasets are continuously annotated with essential information using this metadata schema steadily. This metadata is provided beside the converted data on our data archive and thus enhance its findability and reusability.