Redirected walking for open virtual spaces (How not to walk into walls)

Walking naturally in a large and open Virtual Reality environment is an attractive prospect, but the human mind must be tricked into doing this if the real space is not quite as large.

Bildet kan inneholde: opptreden, scene, begivenhet.

This area has enough research opportunities for several theses.

Imagine that you explore a large and open Virtual Reality environment by walking naturally in a real space. Most of the time, obstacles in the real world will prevent you from doing this in a natural manner, and most VR applications fix this by introducing artificial obstacles in the virtual world. But the human mind can be tricked to forget the direction taken in the real world. This is the approach taken in this thesis.

Imagine you are exploring an extensive Virtual Reality environment by walking in a real space.

You know already that experiencing VR worlds from a chair and using a mouse or some gestures to move around can feel very unnatural and restrictive. For people with a good sense of balance and understanding of their physical presence in space, this can be very disturbing and lead to VR sickness even if the virtual world is displayed without any undue delays. Of course, VR systems like the Oculus Quest and the HTC Vive give you already a lot of freedom to act naturally, but you are still moving in a very constrained space.

Wouldn't it be much more desirable to move your avatar by really moving around in the real world without any spatial bounds?

We have experimented with a much larger space to roam around in during the Third Life project (, but it still required an awareness of the spatial boundaries.

Can we trick the mind to move "freely" when the virtual world is large and open but the real world is constrained?

A VR treadmill (e.g. Kat VR) allows this to a great extent, but it is extremely expensive. A poor man’s solution is real movement in real spaces. But how do we trick the human mind to ignore the physical limitations?

In a set of theses, we want to compare ideas for overcoming this limitation:

  • conscious use of scrolling areas that are indicated by visual or tactile clues
  • unconscious use of rotating VR spaces
  • unconscious use of attention-grabs for VR space rotation

Learning outcome

  • Real-time body and head motion tracking
  • Learning about the human understanding of body pose
  • Manipulation of orientation and position information in VR worlds using Unity or Unreal
  • Objective and subjective assessment of user behaviour and perception


  • Only suited for full-time students. Presence in the lab at IFI is required.

  • IN2140 or equivalent

  • Programming language suitable for implementing real-time interaction in VR (C++ or C#)

Publisert 18. okt. 2021 09:07 - Sist endret 18. okt. 2021 09:29


Omfang (studiepoeng)