Regarding the Virtusphere

[FingerFlinger]

http://www.youtube.com/watch?v=NmpOQZgH ... plpp_video

I recently saw this video about the Virtusphere, which I thought was very impressive, and got me thinking about it in depth.

I got an idea to apply braking or rotational force to the sphere to manipulate the position of the user. With this capability, you can create the sensation of walking up or down an incline of varying degree.



To clarify: When using the Virtusphere with no forces applied, the user will settle at the bottom of the sphere, where it is most level to the ground. But, if we apply a brake to the sphere, and the user continues to walk, they will begin to walk "up" the sphere, and the surface upon which they are standing is no longer level with the ground. By varying the braking force, we can manipulate the user's position in the sphere, which correlates to an angle of incline.

To create a downward incline, the sphere would accelerate in the direction of the user's motion until the user is behind the center of the sphere, and the angle of incline is negative.

Furthermore, you can simulate walking along a hill which slopes off to the left or right, or a spiral staircase(spiral ramp?).

While I don't have immediate plans to build my own Virtusphere, I think it's achievable, and incorporating these kinds of techniques greatly increases the versatility of the device.

Thoughts?

[cybereality]

Yeah, that's a interesting idea. It would probably work to some degree, but I have a feeling there is not a lot of room to work with. If the user walks too far up the side of the ball, then there is the danger of falling over. But it could work, I guess. Hard to tell since I never tried one.

[brantlew]

I think it's an excellent idea. It's hard enough to implement a good omnidirectional motion device - even just for level walking, so most don't even attempt to simulate terrain. But this type of device just might be able to convey a sense of terrain. There would probably be a limit as to how quickly the environmental slope could change (simulating a parking lot speed bump would be challenging) but rolling hills seem doable and would be a significant improvement over an infinite flat plane.

There is another less obvious advantage to having a motorized sphere. I've never used one myself but I have heard that the walking sensation is not realistic because it requires body weight to move the ball. If the rotation was controlled by motors instead of body weight, then a) your center of gravity would stay more centered, b) the stepping surface would be more rigid instead of giving way to your weight, and c) you could use a larger sphere which would make the walking surface a little flatter. All of which should improve the walking sensation.

[cbwan]

I've tried the Virtusphere ( http://cb.nowan.net/blog/2008/05/22/virtusphere-review/ ).

I think your idea could work !

There are currently no motors, and this would complexify the system a lot.
And you need the motor to be able to run in all directions, because the user can be in any direction inside the ball.

Using the motor to get rid of the ball inertia, both when accelerating and decelerating, would also be interesting.
But this also adds lots of complexity:
- As is seen on omni-directionnal treadmills, writing the code for the controller is not easy
- You need to know the position of the user inside the ball. Can be done, but tricky.

cheers

[FingerFlinger]

Yeah, I figured you might be able to drive the sphere with a pivoting wheel on the bottom. I think user positioning would be really tough, I don't have a very good idea for that, aside from measuring the load on the drive motor.

Nice review! Sometime in the future, if I have some garage space, I would like to try recreating the Virtusphere for myself.