Group navigation methods in large-scale collocated VR

Master Thesis
1

Description

In collocated multi-user VR setups, relative positions of users must be aligned in the real and in the virtual world to prevent possible collisions. Recent advances of tracking technologies integrated in VR hardware allows users to navigate virtual worlds by walking at rather large scales – up to 10x10 meters. However, there is an upper limit to the size of a walkable area, defined by the limitations of the tracking technology and of the available tracking area. To enable users to discover the virtual world past such a walkable space, navigation methods are needed. Commercial single-user VR applications often use teleportation to overcome walkable space limitations. However, a simple unmodified teleportation mechanism cannot be used in collocated multi-user VR since it will lead to the loss of alignment between the positions of walking users in the physical and in the virtual space.

The implementation of group navigation methods will be done in the context of the BimFlexi project, although the methods themselves should be independently usable in any application of collocated multi-user VR. The image above shows an example of the virtual environment that will be explored by collocated users.

Tasks

The goal of the thesis is to implement one horizontal and one vertical navigation algorithm for collocated groups of walking VR users. Both navigation methods should maintain the alignment between the physical and virtual positions of users. The developed navigation methods will be evaluated in a user study as part of the thesis.

Requirements

  • Knowledge of English language (source code comments and final report should be in English)
  • Knowledge of Unity 3D
  • Knowledge of networking in Unity is advantageous but not necessary
  • Experience of work on VR projects is advantageous

Environment

The project will be implemented in Unity3D, with Oculus Quest/ HTC Vive used for VR immersion. The networking part will be based on Photon Networking for Unity (PUN 2).

Workplan

The work on the thesis will start with an in-depth overview of existing multi-user and single-user navigation methods. After promising methods have been identified, first prototypes will be implemented and tested. The project will be developed in close contact with the researchers working on the larger BimFlexi project. A user evaluation with several groups of walking users will be designed and conducted as part of the thesis.

It is expected that the implementation of the thesis and the evaluation of the algorithm are finished at least until summer 2022. A bonus of €1000 is possible if completed to satisfaction within an agreed time-frame of 12 months (the details will be discussed with the student in advance).

Responsible

For more information please contact Iana Podkosova.