Author:

Oliver Jung
Supervisor:Prof. Gudrun Klinker
Advisor:Christian Eichhorn (@ga73wuj)
Submission Date:[created]

Abstract

Mixed reality is an fast-emerging technology. It is expanding in the area of entertainment as well as research. However, many driving simulators rely still on conventional LED or projector-based visualization systems. In order to evaluate new designs or interaction systems within such driving simulators, researchers have to produce and install them. In contrast, with mixed reality technology a functional prototype is sufficient. The correct design details are instead added virtually. This master's thesis compares two different approaches of this technology: Virtual reality and augmented reality. The first one replaces the real world completely with virtual content whereas the later overlays only parts e.g., the area of a new interaction device. The focus lies on the influence of the visual inconsistency introduced by mixing real and virtual elements on user experience and specifically the sense of presence. Further, it aims to explore the challenges of installing mixed reality within driving simulators, focusing on object tracking. The implementation is evaluated through a user study within a 6+1 degrees of freedom simulator and a subsequent technical study in a static simulator. As mixed reality headset the Varjo XR-3 is used. In the technical study three marker tracking systems are compared for the use case of tracking the steering wheel: ArUco, CylinderTag and a custom implementation based on the BMW logo. The results of the user study indicate a higher sense of presence in the AR condition and an equal performance and workload in both scenarios. No evidence for an influence of consistent visual environments on the sense of presence is found. In the technical study, the CylinderTag system outperformed the other two systems in terms of accuracy except for conditions with heavy motion blur.

Results/Implementation/Project Description

Conclusion

[ PDF (optional) ] 

[ Slides Kickoff/Final (optional)]