Einsatzmöglichkeiten von VR-Brillen in der experimentellen Verkehrssicherheits- und Mobilitätsforschung

Ensuring sustainable and safe mobility for all road users is an important socio-political task. Studies in virtual reality (VR) have been established as an important tool in experimental traffic research. The validity of their results depends on the realism or immersion of the virtual experience. VR headsets promise a more immersive experience of the virtual world than classic screen- or projection-based simulators. In combination with additional VR components they enable users to move through virtual worlds and to interact with (virtual) objects and people. However, the question arises which combination of VR components, which oftentimes were developed for entertainment-related purposes originally, is most suitable for use as a research tool. Such a valid research tool is especially needed for examining pedestrian behavior, as there is no research tool for this type of behaviour as comparably established as the driving simulator is for studying car drivers‘ behavior. The aim of the present project was to define the requirements for a VR system (i.e., the combination of VR components), for the study of pedestrian behavior, and to provide actionable recommendations for a VR system using VR headsets for the study of pedestrian safety and mobility. A literature and market review was conducted to show the link between the characteristics of VR components (e.g., display features, tracking methods, etc.) and the quality of the virtual experience (e.g. tracking accuracy, realism of visual image, etc.). From this overview requirements for a research instrument were derived. This resulted in a catalogue of criteria that summarizes the main requirements for a VR system in six dimensions (realism, tolerability, data availability and accuracy, universal applicability, experimental economy, and data privacy & safety). Based on a literature review, the research topics investigated so far with VR systems (for pedestrians mainly crossing scenarios in confined spaces) and their technical implementation were described. To identify a VR system that would allow to study pedestrian behaviour in larger virtual worlds, interviews with VR experts (N = 11) were conducted. For each of two scenarios described to demonstrate possible research questions, the experts outlined the most suitable VR system and assessed the extent to which it met the criteria defined. Three types of systems can be derived from the interviews, which differ in the choice of VR components and the feature profiles associated therewith. 1) A stand-alone, location-independent system that is inexpensive, but offers a lower accuracy and a restricted quality of virtual experience –especially with respect to the size of the field of view – than both of the other systems. 2) A laboratory-based system that in contrast to the other two systems, relies on mediated instead of natural locomotion, but thereby manages to limit spatial requirements to normal room size. 3) A high fidelity system that offers both a high quality of virtual experience and natural walking, but is challenging to implement due to demand for space, costs and effort. Based on an overview of important research topics in the field of pedestrian safety and mobility, the scope of applications of these three VR systems was discussed. The overall assessment of the three VR systems assists in choosing the suitable VR system according to the intended purpose and capacity of the respective research institute. Finally, future developments in the field of VR are predicted and possible implications for the results of the present project are pointed out.