VirtuSphere Immersive Virtual Reality

mhzse writes "VirtuSphere provides a mechanical basis for truly immersive virtual reality environments, permitting the user to move about in virtual space by simply walking. The device consists of a large hollow sphere which is mounted on a specially designed platform that allows the sphere to rotate freely as the user walks in any direction. The user wears a head-mounted display, which provides the virtual environment.

Comments

Not great for VR Gaming

[Kainaw]

I've tried this setup twice in the past. It always feels like you are walking up an incline. To make it feel flat, the sphere has to be huge. As it is, the weight of the sphere is difficult to start moving and difficult to stop moving. What is required for a real-life VR environment is a flexible sphere that allows a flat surface for the user to walk on and power-assisted rotation to make it start and stop easily.

Re:I've always wondered how the holodeck worked.

[OverlordQ]

IIRC in one of the Technical Manuals or a book of sorts, it uses tractor beams (or their equiv) to keep people from hitting the walls and using some serious cpu power and force fields in able to refract the light to make people/objects appear the correct distances/sizes/etc.

Coralize! (Let no else's cache evade your eyes!)

[adavies42]

Already down--use http://www.technovelgy.com.nyud.net:8090/ct/Scienc e-Fiction-News.asp?NewsNum=462.

Blocked, linked, and comparison.

[unsupported]

Looks like they all ready blocked referals from Slashdot. Have to copy and paste that URL.

And speaking of URLS, here is the link to the manufactures website.

This thing reminds me of a human version of a hamster ball.

Re:Sgi Cave

[Digital_Quartz]

You can't walk around in the cave. In fact, you'd have to sit in exactly the right spot in the cave, or else the perspective would be out-of-whack at the edges of each of the screens.

Alternative idea

[LinuxPoultergist]

An alternative to the sphere idea is the floor movement idea.

Basically the floor senses where you are going and makes sure to put a moving panel under your foot no matter which way you go.

More info: http://www.newscientist.com/article.ns?id=dn6299

Re:I've always wondered how the holodeck worked.

[kcb93x]

Actually, the holodeck was a combination of light tricks and tractor beams, AND replicators - physical items would actually be replicated, whereas "living" things weren't quite so "physical" IIRC.

HAHA yeah right.

[oGMo]

They had one of these at PAX. Plastic sphere, fairly heavy construction, on some wheels and sensors that allow fairly free rotation. You wore a head-mounted display and had a "gun" peripheral that you could point and shoot. Play time was about 5 minutes and lines were about an hour long.

Guess what? It sucked. Everyone who has considered how to make an immersive VR environment has, at one time, considered sticking someone in a sphere so they can walk around like this. Within 5 minutes, they've also come up with a number of problems with this setup: inertia keeps the sphere going, walking isn't really "flat", you can't run cords into it, and it's expensive and bulky.

I stood in line, figuring they'd come up with solutions to some, or most, of these problems, making it actually usable. They didn't. Stopping and turning was terrible, walking normally took serious focus, and and to top it off, the demo game was unplayably bad: PSX graphics at best, the "which direction is up" calibration was constantly off, it didn't track motion very well, and things just seemed to pop up randomly. And the actual view window was really small. "Immersive" my ass.

This technology isn't worth further investigation until they can prove the above problems are fixable.

Re:Sphere construction how-to suggestions?

[cr0sh]

Unless you have a large room for this, I wouldn't suggest going this route - constructing a 3DOF mechanical gimbal system is not an easy thing to do. It isn't impossible, but it will be difficult, expensive, and big. Furthermore, battery power will not cut it - even if you eliminate the PC powersupply and wire batteries directly, you will be lucky to get an hour or so of running time, if that - which may be OK, but remember - in between sessions you have to wait a long time to charge them back up. These reasons (and others) are why you never see such a system made by large aircraft simulator companies.

These companies instead utilize simpler hydraulic-actuated motion platform systems, which are basically a platform held up by multiple hydraulic cylinders, and upon this platform is mounted the simulator cockpit (sometimes installed inside a large back-projected sphere, as well). For a homebrew system, you will probably not want to mess around with hydraulics. They tend to be messy and corrosive, not something I would want in my house. I would instead look into using larger pneumatic cylinders. These have a drawback in that they tend to be "springy", but they should work out OK. You will also need to set up a position feedback system to these cylinders (and oscillation dampening, etc) so that the platform moves exactly how/where you want it to move.

If you want to find out how to do all of this, there are a few people out there who have done exactly what you are aiming to do - interfacing and building a full flight simulator with motion using a PC - on the internet. Most of them have pictures and details of their system on their websites. You might also look into how various real-world simulators look and work - the history behind aircraft simulation is fascinating, to say the least - it is the birthplace of modern simulation and VR.

Another interesting bit of kit is something called the Rock and Ride - it is basically a two-axis (roll and pitch) gaming chair that you sit in and mount your monitor on. Something like this could be built very cheaply (way cheaper than various asking prices I have seen - plus, I don't think the company that made it is in business anymore).

Also check out Omniscience Futureneering's Joyrider simulator - basically a very homebrew version of the "Rock and Ride".

Before you complain that none of these systems allow full spherical movement, remember that these type systems are similar to what is used commercially. Commercial developers (I am thinking corporations like Honeywell, Rockwell and McDonald-Douglas) decided on motion platform systems not only because of the engineering reasons, but also because they found through testing that full spherical movement wasn't necessary - in a simulation, if you get the physical movement right and in time with the movement of the display, your brain fills in the rest - tilt the cockpit back steeply while performing a loop, and it will feel like a loop! Of course, keep in mind that just like any other simulation, simulator sickness is a very real thing. If you add in a motion platform of any kind, it becomes even more acute of a problem. If your motion doesn't match what is being seen on screen (or in an HMD) - prepare to BARF!!!

Good luck on your project, it is a very ambitious one, certainly!