Solving Treadmill Friction Problem for VR Build

[Cire Venn]

Background: I'm building a treadmill for VR that moves in any direction but I need to figure out if it's worth building or not because of a problem a friend pointed out to me.
Problem: I need to figure out if there's a way to calculate how hard it would be to move this tread by walking on it considering all the forces of your weight and steps are on the rollers which the tread rests between.
https://scontent.xx.fbcdn.net/v/t34.0-12/16507219_1292655057470724_2102035089_n.png?oh=6b537bad9f92d3a97413833e9ab77dc7&oe=5899B160

 

[berkeman]

Background: I'm building a treadmill for VR that moves in any direction but I need to figure out if it's worth building or not because of a problem a friend pointed out to me.
Problem: I need to figure out if there's a way to calculate how hard it would be to move this tread by walking on it considering all the forces of your weight and steps are on the rollers which the tread rests between.
https://scontent.xx.fbcdn.net/v/t34.0-12/16507219_1292655057470724_2102035089_n.png?oh=6b537bad9f92d3a97413833e9ab77dc7&oe=5899B160

Seems pretty simple to build it and try it. Go for it!

 

[rumborak]

A much more interesting question is, how are you creating a treadmill that can go in any direction?

 

[Cire Venn]

@rumborak the picture in my post is a side-view, everything is actually circular, allowing the tread to rotate in any direction around the rollers that encircle it in the way shown.

 

[A.T.]

@rumborak the picture in my post is a side-view, everything is actually circular, allowing the tread to rotate in any direction around the rollers that encircle it in the way shown.

How elastic is the tread?

 

[Cire Venn]

@rumborak the tread has to be flexible, but not stretchy or elastic at all.

 

[Cire Venn]

Seems pretty simple to build it and try it. Go for it!

Ten of those rollers for the design cost 20 dollars, it's going to cost at least 400.0 dollars to build this thing. I can't just build it before knowing if it'll work or not.

 

[berkeman]

Ten of those rollers for the design cost 20 dollars, it's going to cost at least 400.0 dollars to build this thing. I can't just build it before knowing if it'll work or not.

Scale model prototypes are often used for that reason...

 

[Cire Venn]

Scale model prototypes are often used for that reason...

How would I downsize the weight of a running person?

 

[fpsulli3]

@rumborak the picture in my post is a side-view, everything is actually circular, allowing the tread to rotate in any direction around the rollers that encircle it in the way shown.

I'm still not able to picture it. What does the top view look like?

 

[jbriggs444]

How would I downsize the weight of a running person?

http://giphy.com/gifs/addams-family-LLH29xup6RY1W

 

[sophiecentaur]

Ten of those rollers for the design cost 20 dollars, it's going to cost at least 400.0 dollars to build this thing. I can't just build it before knowing if it'll work or not.

Before going any further you need to consider the structure and topology of the surface you walk on. For it to move in any direction, it needs to be continuous in all directions. A sort of deflated sphere with balls inside it, resting on the drive mechanism; more balls or rollers. It's hard to visualise how to drive and control it. It would also need to slope in all possible directions.
You could have a simple low friction surface and servo'd restraints on the user's feet to give the illusion of walking forward without the base actually moving

 

[jbriggs444]

The picture I have in mind is the same as that of @sophiecentaur -- something like an inflated track ball, flattened on top. The more elastic the fabric, the bigger the flattened area can be relative to the complete sphere.

 

[Cire Venn]

The top view is a circular closed tread surrounded by rollers. The inside along the edge has rollers too.

 

[sophiecentaur]

The top view is a circular closed tread surrounded by rollers. The inside along the edge has rollers too.

B ut if it is circular, what do the rollers roll on? You have an enclosed surface with rollers inside it or an infinitely stretchable membrane?
Have you tried a a 3D picture of your design?

 

[Cire Venn]

Dude, look at the picture in the post, the rollers inside the round track roll on the rollers underneath and around it.

 

[DaveC49]

You have a problem with the translation from a 2D section to the 3D object. The distance from the centre of your device to any point on the periphery of your tread has to be constant which means that the surface of the track has to be spherical, not a flattened sphere, if it is inelastic. A sufficiently elastic material may be able to be flattened out into an approximation of a disk and be deformable enough to roll. One silly question, in the latter case how do you plan on getting your roller mechanism onto the inside of your tread since it has to be either a closed sphere or a rotation of an ellipse around the minor axis. Any join in the tread put in place after the tread is placed around the rollers is unlikely to have the elasticity of the raw material so you would not achieve the even deformation for your tread to roll readily in any direction. Someon suggested trying a 3D solid modelling package. I think you would find it extremely instructive to attempt to model your device..

 

[DaveC49]

How would I downsize the weight of a running person?

Why would you need to? The point is to build a scale model of the actual device. You can use a couple of fingers to simulate your running person. You have much more serious geometric problems with the shape of the tread than anything else, particularly if as you suggest it is an inelastic material.

 

[sophiecentaur]

Dude, look at the picture in the post, the rollers inside the round track roll on the rollers underneath and around it.

We can all see that but what keeps the rollers up if, as I understand, the walking surface needs to move in all directions? The membrane has to get underneath everywhere.
You might be able to use a tray full of extremely small balls and no membrane. That would allow movement in all directions. Steel balls and strong magnets to keep them in place on the tray?

 

[sophiecentaur]

. . . . or a massive sphere to walk on - >2m radius.

 

[Khashishi]

We can see a side view of the ribbon, which is loosely pulled around the coasters. But we can't see how the ribbon is shaped in the third direction, in and out of the screen. You need to provide another view so we can figure out what the 3D shape is.

 

[A.T.]

. . . . or a massive sphere to walk on - >2m radius.

Or walk inside of it:



https://www.researchgate.net/publication/220425224_Cybersphere_The_fully_immersive_spherical_projection_system

 

[sophiecentaur]

Or walk inside of it:



https://www.researchgate.net/publication/220425224_Cybersphere_The_fully_immersive_spherical_projection_system

Shame it's too big for a lad's bedroom - in amongst the drum kit and the Hi Fi speakers.

 

[rumborak]

You might be able to use a tray full of extremely small balls and no membrane. That would allow movement in all directions. Steel balls and strong magnets to keep them in place on the tray?

Yeah, the "thousands of balls" idea is the only feasible one I think. If you sandwiched the ball bearings between two plates that have regularly spaced holes of a diameter slightly less than the ball bearings, they would stay in place. Then, you place an actual treadmill under it which however can rotate. That way the balls will rest on the treadmill, thus transferring the motion,and the rotation of the treadmill will achieve the change in direction above.

 

[sophiecentaur]

Yeah, the "thousands of balls" idea is the only feasible one I think. If you sandwiched the ball bearings between two plates that have regularly spaced holes of a diameter slightly less than the ball bearings, they would stay in place. Then, you place an actual treadmill under it which however can rotate. That way the balls will rest on the treadmill, thus transferring the motion,and the rotation of the treadmill will achieve the change in direction above.

I was thinking that the angle of the balls (if they were marked suitably) would show the movement of the feet across them. An optical system for motion recognition and measurement would allow the application to map the progress of the user through the virtual landscape. That would be very smart and have no moving parts - apart from the balls, which would have to be free to move against the soles of the feet.
Magnetic balls could return to a vertical direction when foot is removed and could be black on the bottom. The white upper bit would only show when the feet were moved. Pressure of a stationary foot could be detected by the depression due to weight showing an equatorial coloured ring.