Moving about in VR: Redirected Walking

There are different ways of moving about in a virtual world. The most intuitive and most natural way is simply through the movement of your own body. Thus, you can explore the virtual environment just as you would in the real world. The obvious disadvantage, though, is that the virtual World is constrained by the space that is physically available to you at the time. In this article, I will list some techniques of representing movement in VR and explain why Redirected Walking is a particularly interesting example among them.

 

Since VR came into existence, the various techniques for creating movement in the virtual world have been continuously improved and there are many different variations. Every one of them has certain advantages and disadvantages and is more or less suitable for various applications. While in the past, the greatest challenges in VR were hardware-based (field of vier, resolution, etc.), the issue with locomotion can be solved largely via the software. The goal is to give the users a feeling of movement that is as realistic as possible, while taking into account the limited physical space and preventing the occurrence of motion sickness. Let’s take a look at some of the more widespread techniques for representing movement in VR.

Smooth Artificial Locomotion

Here, you move through the VR space like you would in a traditional console game, by using a trackpad or thumbstick.

Flight

Stefan already mentioned this in his recent article on here. This method was developed for the game Eagles Flight and the direction of movement is controlled by the player’s line of vision.

Teleporting

Teleporting is another very simple and intuitive method. The player looks or points in the direction he wants to move in and presses a button on the controller in order to jump to the desired location.

Armswinger

For this method, a motion-tracked controller is needed. You move around the VR space by imitating the arm movements you would make while walking in the real world, but your feet remain planted on the ground. The faster you swing your arms, the faster you walk in the virtual world. A variation of this method allows for swinging only one hand while the other remains free for using weapons and other interactions.

Bobbing-Walking

Like in the armswinger method, natural body movements exhibited while walking or running are used as input. Here, it is the head, which usually moves up and down when we walk or run. Thus, you need to imitate this movement in order to move forward when this method is used.

Redirected Walking

Since the available physical space around you is usually limited, most methods are so-called in-place methods, because you remain sitting or standing in one location. However, it would feel much more natural to move through the virtual world the same way as through the real one. The problem in this case, though, are physical boundaries and a potential risk of injury through unexpected obstacles (walls, chairs, …) 😉

A simple solution would be ‘walkabout locomotion’. Here, you can manually turn the virtual environment. This means that once you’ve arrived at one end of a room, you ‘freeze’ the environment and turn it so that you can move freely again without actually having to turn in the game. This method is a simple form of redirected walking and can be quite suitable for some applications.

However, so-called “behind-the-scenes” processes are much more user-friendly and are ideally not noticed as such by the users. Originally, this technique was inspired by the fact that a blindfolded person will always walk in a curve, even if they are asked (and are trying) to walk in a straight line. Applied to VR, this means that you are trying to manipulate the users so that they change direction without noticing or getting motion sickness. Extreme example: you think you’re moving straight along a long corridor, but in reality, you are walking in a circle. This effect can be achieved by rotating the virtual scene around the user, ideally, without them noticing. We humans mainly rely on vestibular, visual and auditory perceptions for balance and orientation. We also use these senses in order to check whether we or the objects around us are currently moving. It’s the goal of redirected walking to maintain these sensory inputs as continuously as possible, so that the user doesn’t notice the environment’s movement. Of course, there are also some disadvantages to redirected walking. It needs a lot more physical space than the other methods described above. Further, the user path in the virtual world might have to be predefined.

Redirected walking is an interesting concept with a lot of potential. NVIDIA, Adobe and researchers at the Sony-Brook-University only recently presented a further redirected walking system. In this, the virtual scene is manipulated while we are blind for a very short amount of time. More accurately, it uses the saccadic eye motion, which Stefan also mentioned in his article. Saccades are rapid and unconscious eye movements which occur when you look from one part of the scene to another. When you are following a moving object with your eyes and even when you are focusing on a certain spot, your eyes move around rapidly and in between focusing different objects, we are essentially blind. This blindness only lasts for a few milliseconds but can be used to minimally shift the virtual scene during this time. Apparently the system can also react to dynamic changes (like, for example, a person walking into your path) and guide the user around an obstacle.

It will likely still take some time for this system to actually enter the market, though. And I, for one, am very curious to see what other creative methods of implementing movement in VR will be developed over the next few years.