Idea for an open source locomotion platform! ("LocoVR")

[lowv]

**********
DISCLAIMER: I'm about to launch this as a free open-source project, built from available parts with only a drill press and one or two welds. Please contact me if you're interested to participate...

Hi,
I have been lurking around here for a long time, reading up on all the existing options for movement controllers and locomotion devices. Sitting down on a chair vs using an ODT seems to be a frustrating choice, full of compromises (immersion vs comfort vs price and space requirements...). Yesterday it occured to me that there might be an intuitive solution that is affordable for regular users...
I present you: the "LocoVR" (thanks for the name suggestion!)

It is basically a cross between a small bar stool and a kind of oversized joystick (small deflection [~10°], stiff and tilt resistant) with a swivelling top part (interchangeable for free 360° rotation or spring loaded incremental turning). The shaft is set to a length that allows your feet to barely touch the ground, supporting most of your weight in an upright position! You get on top of the seat with your hips behind the padded crossbar, turn around and lean slightly into the direction you want to move. The axial deflection itself is constrained to a relatively small and safe degree, while your upper body will tend to bend over naturally to gain additional inclination (stand up and take a moment to try and visualize this. Keep in mind that your weight will be supported!).

Advantages:
1) Shifting your balance will fool your vestibular system into thinking that you're actually moving . In this way (and by using the seat orientation as an input) it becomes possible to decouple looking and walking, just like with an omni-directional treadmill.
2) Since you're sitting erect, there is no disconnect between your avatar's view and your lowered position in a chair. I found this to be quite annoying and immersion-breaking, a major disadvantage of playing in a sitting position.
3) It's comfortable: I imagine that running around on a passive treadmill is great, but could actually discourage you from longer gaming sessions once the novelty wears off.
4) It's stowable: breaks down into 4 parts.
5) It's affordable: With good sourcing, it should cost around 100-150 Euro





The recessed end flange acts as a tilt resistant bearing (adjustable by the contact area) in order to support a stable neutral position under one's body weight.

[WiredEarp]

I've actually done something very similar to this, and many other experiments with what I call hybrid movement - where the turning is limited and turns your character in a manner similar to using a joystick.

The version that was similar used a swivel chair, with the back castors raised a bit (they were hot glued to my punching bag stand). Strong bungies (edit: actually bike inner tubes) went to the bag stand to provide autocentering and keep the amount of movement to about 10 degrees each side. A TrackIR passive reflector tracker was hot glued to the underside of the chair, and the trackir sensor mounted on the bagstand. This let me then use swivel action to turn my character in my VR testbed. The angle of the chair helped with intuitive turning. I was using mouse sensors on shoes to provide forward movement detection. It would have been improved with the barstool idea slightly, since my swivel chair could not gain sufficient height to duplicate full standing.

I actually found it quite satisfactory for myself, but it wasn't immediately intuitive for everyone, just like all the hybrid concepts I tried. It was certainly better than using a joypad however. My other main hybrid system of a turning pedal combined with an exercycle was better feeling again however, since it allowed more natural feeling walking forwards (despite being a rotary motion). My current system, which I unfortunately cannot talk about as I believe it is good enough to have commercial applications, is more natural/intuitive, which I feel is important. The big problem with hybrid systems etc is you need to train yourself a little to use it (although, proper chair designs could minimize this) and that the more you make it like actual walking (which is really what is required) the easier it is to forget your training and just trying to walk (unsuccessfully). Its the uncanny valley' issue, applied to locomotion.

Oh yep, dont call it 'VR Stool' - that sounds like something for Toilet Simulator . I'd go with SpinVR off the top of my head

[Zoide]

My current system, which I unfortunately cannot talk about as I believe it is good enough to have commercial applications, is more natural/intuitive, which I feel is important.

Soooo..... When might we hear some more about your system?

[1karl1]

I like the look of this and would be interested . Looks a great solution for those like me who like the idea of some standing VR but not the whole exercise bit . It's ease of use looks a good selling point and would imagine it won't be a pain to move about . The name could do with changing as not only the word "stool" but the word "arse" is used when pronouncing "VR Stool" . LOCO-VR maybe idk but best of luck with this

[lowv]

I've actually done something very similar to this, and many other experiments with what I call hybrid movement - where the turning is limited and turns your character in a manner similar to using a joystick.

The version that was similar used a swivel chair, with the back castors raised a bit (they were hot glued to my punching bag stand). Strong bungies (edit: actually bike inner tubes) went to the bag stand to provide autocentering and keep the amount of movement to about 10 degrees each side. A TrackIR passive reflector tracker was hot glued to the underside of the chair, and the trackir sensor mounted on the bagstand. This let me then use swivel action to turn my character in my VR testbed. The angle of the chair helped with intuitive turning. I was using mouse sensors on shoes to provide forward movement detection. It would have been improved with the barstool idea slightly, since my swivel chair could not gain sufficient height to duplicate full standing.

Sounds interesting. Do you have any photos? It's seems like quite a different concept though. Did your translational movement also involve leaning? Because I really believe this to be a great compromise between static play and full locomotion.

[WiredEarp]

Sounds interesting. Do you have any photos? It's seems like quite a different concept though. Did your translational movement also involve leaning? Because I really believe this to be a great compromise between static play and full locomotion.

Unfortunately, I don't have any photos (been told I should really start taking some and recording what i've tried), although I could put the chair back in the situation it was in and take one I suppose.

It didn't involve a massive amount of leaning, although having the rear of the chair elevated (so the seat is on a slant) naturally gives a bit of lean when you turn (and turn when you lean), which does help with both the intuitiveness and the feel. If I'd had some blocks to securely support it, then I'd have raised it another 6" which would have helped quite a bit I think.

Slightly related (although no lean), there is also an arcade VR system (in Europe somewhere AFAIR) that uses a 360 degree spinning stool to turn. I believe it has a button or joystick to move forwards.

@ Zoide:

Soooo..... When might we hear some more about your system?

Probably about 6-8 weeks before I'll be willing to show it off.

[IanBruce]

If all you want to capture leaning force, I might suggest using as little as 4 piezoresistive force sensors in a seat cushion.

[lowv]

If all you want to capture leaning force, I might suggest using as little as 4 piezoresistive force sensors in a seat cushion.

That would be a very neat solution! I haven't thought about that. Trying to visualize this, my only concern would be that the "feeling" could have a missing component to it, i.e. there wouldn't be the same "give" which I think is very helpful for sensitivity.

[IanBruce]

If all you want to capture leaning force, I might suggest using as little as 4 piezoresistive force sensors in a seat cushion.

That would be a very neat solution! I haven't thought about that. Trying to visualize this, my only concern would be that the "feeling" could have a missing component to it, i.e. there wouldn't be the same "give" which I think is very helpful for sensitivity.

"Give" could be added by using a 2-element seating pad fixed by a centrally positioned floating mount. You'd just use a layer of foam between them to dampen the motion. It's also not hard to imagine using a torsional sensor (fixed) or an inertial accelerometer to capture swivel axis. I shall call it ButtMouse™ in honor of Richard Gere.

[usb247]

combine this with a stair stepper that leans and rotates?

Basically looks good though, anything to get from being confined to a treadmill.

[1karl1]

Foot position while using this to maintain balance would be one foot slightly in front of the other depending how much of a lean angle is needed to operate as side by side it would be easier to fall over . Turning could be analogue in a similar way to how a steering controller works with a 90-180 degree rotation .

[lowv]

Foot position while using this to maintain balance would be one foot slightly in front of the other depending how much of a lean angle is needed to operate as side by side it would be easier to fall over . Turning could be analogue in a similar way to how a steering controller works with a 90-180 degree rotation .

You probably shouldn't imagine it to work quite like a joystick. The translation is much more nuanced, with most of the bending coming from the upper body, in a natural way. So there shouldn't be any problem with balance... But yes, setting one foot forward would add a nice touch of extra immersion.

Turning could be analogue in a similar way to how a steering controller works with a 90-180 degree rotation .

That's a very good idea! One could even think of a mechanism (or simply interchangeable top parts) that switches between that kind of control and full 360° rotation. This way, it would be possible to play games that are not optimised for VR, or people could simply choose based on preference...
I've edited it into the original post, thanks.

[NeoTokyo_Nori]

Advantages:
1) Shifting your balance will fool your vestibular system into thinking that you're actually moving (TRY it!). In this way (and by using the seat orientation as an input) it becomes possible to decouple looking and walking, just like with an omni-directional treadmill.
2) Since you're standing erect, there is no disconnect between your avatar's view and your lowered position in a chair. I found this to be quite annoying and immersion-breaking, a major disadvantage of playing in a sitting position.
3) It's comfortable: I imagine that running around on a passive treadmill is more of a novelty that, once it wears off, will actually discourage you from longer gaming sessions. I think people seriously underestimate this aspect...
4) It's stowable: breaks down into 3 parts.
5) A commercial product could probably be built for around 100-150 Euro

HI, I like the idea of having affordable alternatives to the OD Treadmills, about which the more I think about the more I realize that -
once you have walked or run for any longer than the < 60 second clips we have seen of people on them, you will realize that "oh, I have to continue running and walking for the entire time that I want to play the game" which for any gamer is going to be 30 minutes to 1 hr at minimum, and much longer usually.

This totally defeats one of the elements that make gaming (such as fps) enjoyable, which is that you essentially gain the super-human ability to run around at high speed without ever running out of breath, or getting tired. Almost all the initial buyers for the ODTs most likely do not have the physical ability to run around for extended amounts of time, and neither do I.

Therefore I doubt their effectiveness as (fps) game controllers in the long term, but they may work as a physical training related controller. And most physical training equipment will take commitment to be keep using it, and not collect dust.

To get back to the topic, the "VR stool" name you suggested at first, would have people wondering if they need an additional controller for "VR bowel movements" so yeah, we do need another name

Although I think your design is a good starting point, I imagine that the leaning of the entire body, and not just the upper torso that is necessary to control it could be a potential problem - due to the fact that you are assuming everyone who uses it is physically coordinated enough to NEVER loose their balance and topple over while wearing HMD.
Also the bottom base will have to be something very heavy - probably solid iron etc. - to be stable and sturdy enough so that it too, NEVER allows the person standing on it to tip over.

When you consider the safety requirements of a commercial product, these could be large obstacles to overcome. If you want to go with "leaning" as the control scheme, then it may be necessary to restrict it to upper body leaning...

Anyway, its interesting to think about these things

edit; is there some reason you cannot have the people SIT on the controller, like you know, a stool?
then there would also be far less of the "toppling over" problem, although maybe not entirely....

[lowv]

once you have walked or run for any longer than the < 60 second clips we have seen of people on them, you will realize that "oh, I have to continue running and walking for the entire time that I want to play the game" which for any gamer is going to be 30 minutes to 1 hr at minimum, and much longer usually.

This totally defeats one of the elements that make gaming (such as fps) enjoyable, which is that you essentially gain the super-human ability to run around at high speed without ever running out of breath, or getting tired. Almost all the initial buyers for the ODTs most likely do not have the physical ability to run around for extended amounts of time, and neither do I.

I totally agree. Watching these videos is very tempting so it's understandable that people fall for the hype without considering the practical aspects...

To get back to the topic, the "VR stool" name you suggested at first, would have people wondering if they need an additional controller for "VR bowl movements" so yeah, we do need another name

I picked up 1karl1's suggestion and changed it to "LocoVR". Though ultimately that's not up to me. As I said, I have no interest in marketing it myself. Just want to spread the idea...

Although I think your design is a good starting point, I imagine that the leaning of the entire body, and not just the upper torso that is necessary to control it could be a potential problem - due to the fact that you are assuming everyone who uses it is physically coordinated enough to NEVER loose their balance and topple over while wearing HMD.

The allowable inclination would be constrained to a safe degree. Sensitivity should still be good due to the interplay of high resistance vs high force from your hips.



Can anyone propose a good way of spreading this concept around? Who might be interested in adopting this? Again, I do not seek any commercial gain from this, I'm just interested in it myself and believe that others might also be, assuming that it works as imagined...

[NeoTokyo_Nori]

Can anyone propose a good way of spreading this concept around? Who might be interested in adopting this? Again, I do not seek any commercial gain from this, I'm just interested in it myself and believe that others might also be, assuming that it works as imagined...

My suggestion would be to consider making it an Open source hardware project.
http://www.oshwa.org/definition/

Excerpt from site:
Open source hardware is hardware whose design is made publicly available so that anyone can study, modify, distribute, make, and sell the design or hardware based on that design. The hardware’s source, the design from which it is made, is available in the preferred format for making modifications to it. Ideally, open source hardware uses readily-available components and materials, standard processes, open infrastructure, unrestricted content, and open-source design tools to maximize the ability of individuals to make and use hardware. Open source hardware gives people the freedom to control their technology while sharing knowledge and encouraging commerce through the open exchange of designs.

I am actually considering doing so with one of my projects

[IanBruce]

I'd suggest you Google "standing stool" or "wobble stool" and look at the image results. That may spark some additional ideas.

[lowv]

My suggestion would be to consider making it an Open source hardware project.
http://www.oshwa.org/definition/

Thanks. I will probably do that. What kind of repository are you going to use for your project (btw, what is it?). Github and others aren't meant for hardware, and a wiki isn't ideal either...

I'd suggest you Google "standing stool" or "wobble stool" and look at the image results. That may spark some additional ideas.

Interesting, thanks.

[NeoTokyo_Nori]

Thanks. I will probably do that. What kind of repository are you going to use for your project (btw, what is it?). Github and others aren't meant for hardware, and a wiki isn't ideal either...

Hi,
If you look at sparkfun products which are mostly open source hardware,
it appears that they host their design files on their own site (a repository),
but also on github, since you can obviously store digital files of hardware designs.
https://www.sparkfun.com/products/12651

I dont know how much open hardware there is currently that are not electronic pcb related things,
but I dont think there is a reason it cant be a controller widget either...

[lowv]

I finally decided to bite the bullet and do a proper build according to my latest revision (see first post). It will be based on unicycle parts and simply measure the tilt via an accelerometer (very cheap), while a 360° servo pot will take care of the orientation.

[NeoTokyo_Nori]

I finally decided to bite the bullet and do a proper build according to my latest revision (see first post). It will be based on unicycle parts and simply measure the tilt via an accelerometer (very cheap), while a 360° servo pot will take care of the orientation.

Great, a man should put his money where his mouth is.

A unicycle seat ! interesting, I would have never though of that.
Being able to sit on the device, solves a lot of the potential loosing-balance, and tipping-over issues,
as well as weight of the base, so I think this is the way to go.

I think the angle to lean should be limited by a hard (metal or something) stopper, and not rely on just the spring strengthm, to stop it, for safety.

Just a note, you are mixing up servo and potentiometer, which are 2 different things, although a servo does contain a potentiometer inside it. confusing yes .
A simple potentiometer should be sufficient for this case, and they are less an $1 compared to at least $10-$15 for a servo.

[lowv]

A unicycle seat ! interesting, I would have never though of that.
Being able to sit on the device, solves a lot of the potential loosing-balance, and tipping-over issues,
as well as weight of the base, so I think this is the way to go.

I would have never considered such a device unless you could comfortably rest on it . Someone else told me about the unicycle idea. It's not just the seat, but the whole seatpost assembly minus the ball bearing, which can be adopted without having to struggle with custom parts.

I think the angle to lean should be limited by a hard (metal or something) stopper, and not rely on just the spring strengthm, to stop it, for safety.

It's all accounted for: That's what the acetal guide rods inside those springs are supposed to do.

Just a note, you are mixing up servo and potentiometer, which are 2 different things, although a servo does contain a potentiometer inside it. confusing yes .
A simple potentiometer should be sufficient for this case, and they are less an $1 compared to at least $10-$15 for a servo.

Yeah, it's misleading, I now realize that. It's kind of a synonym for heavy duty 360° pots. The 1$ ones normally don't turn continuously...

[lowv]

MODS, could you please move this thread into the VR/AR Develpment subforum?

[cybereality]

This is the correct section. It was made specifically for ODTs and other similar devices.

[WiredEarp]

lowv, looking forward to seeing what you come up with. Us kiwis are big fans of 'lets give it a go' as an experimentation technique