Walking in VR [Long, Long, Long Post]

"brantlew" wrote:
...

Oh I hadn't seen the Wizdish before. It looks like it would have been pretty cool. But how does it differentiate between walking backwards and walking forwards? I also couldn't find any information about what sensors it uses, other than an unverified mention of IR depth-sensors. I'll add it anyway.
I'll not add the Virtuasphere, just because it's definitely not a consumer device. When you have to email the company to get a quote, it's not going to be accessible to the average consumer. ;)
And yeah, I agree that using a large area and simply detecting real motion and interpreting it as virtual motion is the ideal solution. However, it defeats one of the largest advantages of VR - that is, it frees us from many of the restrictions of the material world! Plus not everyone has a square kilometre of flat ground in their backyard. I'll put it in as well though, for completeness' sake. ;)
You might want to check out http://zerolatencymedia.com/ - it's a virtual reality setup in Melbourne (there are probably similar things elsewhere in the world) that uses this kind of 1:1 walking input for locomotion.

"sutekiB" wrote:
...

See for me, a passive ODT ala Omni is already too expensive, and the movement-limitation imposed by the harness isn't acceptable. I would be able to look past both of these though, since the expense will go down as more and more competition arrives, but the harness is a completely necessary component of these ODTs. I think they're a good start, since if nothing else they'll make sure everyone's keenly aware of the necessity for a good VR Walker.
I do agree that they're head-and-shoulders above any motorised locomotion input device, since the momentum is always going to be a huge problem. Even with predictive algorithms, there are always going to be relatively common use-cases where there is a sudden unpredictable change in the user's direction (e.g. going from sprinting to a sudden stop) that will be impossible for any system to keep up with.

"lmaceleighton" wrote:
...

That sounds interesting - if the surface the user is walking on is just carpet, how are you capturing the walking motion?
And more importantly for me, how are you removing the user's weight from the surface? Are you planning on using a harness like the Omni and Virtualizer?

"latearrival" wrote:
...

Looking over their website, it seems to me that they have big plans, but no details for how to make them come to fruition. I've been thinking myself about having a locomotion device that sticks on your feet, kind of like Stompz. However I lost confidence in project VR Walker (I wish they hadn't stolen my name D:) when I read this:
"It's motorized and controlled with an array of sensors. The sensors tell the shoe modules what to do to keep you from travelling through space as you go through the natural walking motion."
That implies to me that it will be a system kind of similar to this http://www.gizmag.com/go/5811/, but aimed at the consumer VR market. The problem with this is the same problem that was mentioned by sutekiB - it just doesn't seem possible that any motorized device like this will be able to track and respond to movement fast enough to keep up with sudden changes in speed or direction. Plus the vagueness of the paragraph and the lack of any further information, images or videos on the site hint to me that the project is just an idea with no testing, prototypes or money behind it.
I'd love to be proven wrong though. :D