Infinadeck is a new entrant into the omnidirectional treadmill space for VR use. The device allows a user to walk completely naturally in any direction. Unlike other VR treadmills that have sprung up recently, the Infinadeck is an active, rather than passive, solution to the problem of VR locomotion.

Bats? Omnidirectional Treadmills? It’s All Just Engineering.

Having seen what, to my eyes, appeared to be a very complex piece of machinery, I would have never thought that the Infinadeck treadmill was designed and built by one guy—in his spare time. George Burger, the inventor of Infinadeck, is the also Director of New Product Development for Louisville Slugger, the baseball bat brand. He designs the machines that make the bats.

When I asked him, quite amazed, how he managed to build this monstrous device on his own, he stared at me puzzled, like I’d just asked him how he managed to tie his shoelaces. From the looks of the individual belts running along a tank-tread-like conveyor, I had assumed that each belt was powered by an individual motor. Burger went on to explain that the Infinadeck omnidirectional treadmill is powered by just two motors—so much for my engineering intuition.

In reality, the device consists of just one belt that is looped in such a way that it comes out straight on top, giving the appearance of many belts that run perpendicular to the direction of the conveyor. Using one belt instead of many has two huge benefits. First, the belt can be powered by one motor instead of multiple, hugely cutting cost. Second, a single belt provides a constant speed across the entire system, with no need to sync motor RPM—very important for safely walking on this massive machine. By combining the motion of these perpendicular systems with the vector of the user’s direction, Infinadeck lets you walk in any direction using a single platform.

SEE ALSO
Vision Pro Preview: Early Thoughts on My Time Inside Apple's First Headset

Walking on Infinadeck and Challenges With Active Omnidirectional Treadmills

infinadeck omnidirectionl vr treadmill

I got to walk on the Infinadeck at the SVVR Conference & Expo last week. For now, it lacks the ability to detect a user’s direction and speed. Instead, Burger manually controlled those values using two makeshift dials. Using my arms I indicated to him which direction I wanted to turn, and he adjusted the vector slowly as I rotated.

When I wasn’t rotating, it worked very well. No matter what direction I was facing, walking was natural and seamless. Manual rotation in symphony with Burger was a bit precarious, a few times I nearly lost balance (I can’t imagine it being too fun to fall onto this beast). At the conference there wasn’t any safety ring around the user, but this is apparently in the works. A safety rig to prevent the user from falling will definitely be necessary for virtual reality use.

Other VR treadmills like the Virtuix Omni and Cyberith have taken a passive approach to omnidirectional motion. They use low-friction surfaces to facilitate locomotion rather than moving parts. The passive approaches reduce size, cost, and complexity, but may not be as natural as walking on a moving surface.

The usefulness of Infinadeck will come down to how well the user’s direction and speed can be detected and compensated for. If it’s off by just a bit, losing balance becomes a very real possibility. Not being able to see your legs in a virtual reality environment (occluded by a VR headset) would make this even more difficult.

Momentum could be another problem for the prototype. Due to its heft, there’s no way I could see the prototype working for a jogging pace followed by a quick stop; the parts likely wouldn’t be able to stop fast enough to keep the user from losing balance. However, Burger says that he built Infinadeck with scalability in both size and manufacturing in mind. A production model, he said, could be much lighter, and perhaps more responsive. Noise was also a problem that Burger said would be improved in a production model which would use plastic gears instead of chain sprockets.

SEE ALSO
Ubisoft "disappointed" in 'Assassin's Creed VR' Performance, Future VR Investments on Hold

Assuming the parts can stop fast enough, sensing user speed and direction will be vital. A solution as laggy and jumpy as the Kinect could never work. Infinadeck will need something with a lighting fast sampling rate and incredible accuracy if they want to preserve completely natural walking motion.

That’s not to say that Infinadeck wouldn’t be great for some games in its current form. So long as users don’t expect to be running at full sprint in their favorite FPS, Infinadeck could provide the perfect platform for virtual tourism, and slower experiences like the excellent Dear Esther (2012) which has the player walk through a beautiful environment with a guided narrative—no running needed.
Newsletter graphic

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. More information.


Ben is the world's most senior professional analyst solely dedicated to the XR industry, having founded Road to VR in 2011—a year before the Oculus Kickstarter sparked a resurgence that led to the modern XR landscape. He has authored more than 3,000 articles chronicling the evolution of the XR industry over more than a decade. With that unique perspective, Ben has been consistently recognized as one of the most influential voices in XR, giving keynotes and joining panel and podcast discussions at key industry events. He is a self-described "journalist and analyst, not evangelist."
  • monographix

    I am amazed it uses only two motors. It takes a genius to make something so complicated work in a simple way.

    “Infinadeck will need something with a lighting fast sampling rate and incredible accuracy if they want to preserve completely natural walking motion.”

    Wouldn’t PrioVR actually be perfect for this task? As for losing balance, makers of similar treadmill managed to somehow solve one of the natural motion problems, but how many are still left is a mystery.

  • Tim Suetens

    I don’t see this thing having a future. Whoever believes in this is completely out of touch with reality.

    People don’t want the Rift because it makes them look dorky; and you think they’ll install a 1000 pound machin ein their house/apartment just for a few shitty VR demos? Hahaha.

  • Roy

    Hum, I wonder how well this would work if it was passive as well.

    If you’re going to introduce a harness, like the Virtuix Omni or Cyberith then would this not work as a better surface for walking on and just have move with the pressure from your movement rather than trying to predict it.

    Surely there’s a problem that could come from wearing a static harness then putting your foot out in front of you like a step but placing it with the intention of stopping.

    I’m also keen to know how this actually performs with a VR headset.

    I know you mentioned that your balance felt funny even when the motion was a little out of sync but is that partly because you’re sight reference is a static world rather than the virtual world that you’re feet are moving around in?

    Also, I wonder how accurate movement detection needs to be. Wouldn’t one possible approach be that rather than the system trying to accurately trace your steps it simply just tries to move you to the middle of the device.

    If the system used a Kinect (or even a webcam) simply to detect your position on the device it could simply move you very steadily to the middle. Again, with the real world as a reference this would likely be quite disorientating, but with a VR headset on, if you can’t feel the movement from below and the world isn’t moving then there’s no disconnect and you’d never know it’s happening… possibly :-)

  • Psuedonymous

    “the device consists of just one belt that is looped in such a way that it comes out straight on top, giving the appearance of many belts that run perpendicular to the direction of the conveyor”

    So there IS a functional difference between the Infinideck and existing 2-axis treadmills!

  • Druss

    Love the idea, but just having build the mechanics alone isn’t going to cut it. The only way I see this working perfectly (and that is the only way it will have any use at all) it pressure+motion+positional sensors in the shoes (or maybe a sole) that detects how you are distributing your weight etc. and then you need an advanced algorithm that can adapt to the individual and anticipates their speed and direction. That will be crucial in detecting sudden changes in direction, speed, etc. that you need in fast paced games. Nobody is going to buy this just to walk around slowly, although that will of course be a big part of it.

    • Druss

      Sorry for the double post (please add editing functionality to posts). I just had to add that I think this guy is a fantastic engineer and this idea deserves to get picked up by a big company. Oculus themselves would be ideal, with Facebook backing them I think they should start expanding towards all facets of VR, not just the visual aspect. Them also developing a UD-Treadmill and a way to track hand motion could lead to a VR platform that works together seamlessly and one that should be a lot cheaper than buying all the things from different, smaller companies. Hand tracking is something I NEED in VR, it’s supposed to mimic reality and I (mostly) don’t press buttons to open doors in reality. I don’t think I need to rehash the ‘VR+UD-Treadmill == healthy gamers’ argument, but the better the treadmill the more people will use it.

    • srfe

      “and then you need an advanced algorithm”

      Gee ya don’y say.