Post by Frisbone on Mar 1, 2013 10:49:26 GMT -5
Basically I'm looking at a new design for an omnidirectional treadmill. Initial thoughts are to have a bed of beads on bars forming a circular platform. The bars would be on bearings inside a larger ring (although could be rectangular?). The beads would be held down by a mesh that is supported through the cracks that allows the beads to protrude slightly.
purpose of the beads would be to allow free movement of a person that is strapped in at the waist. The wait would be mostly supported by a super structure with just a small amount applied to the bead bed.
Our goal would be to be able to detect average direction of motion and distance from the motion of the beads. This would mean we would have to have some special beads that could be tracked. One possibility could be that we imbed small magnets in beads (so plastic would be best). If we had some matrix underneath that could detect a disturbance when the different poles passed over it we could obtain a picture rotation rate. Actual rotation direction might be trickier. Perhaps if the matrix underneath was small enough compared to the magnet, direction could be determined (X/Y).
So if we actually could get rotation direction and speed from the matrix underneath we could average them together to get a sense of what the subject was doing. If they are strapped in fixed fashion then movement can correlate to using a movement joystick (forward, backward - Y, strafe left/right - X.
Balance may become an issue though as the body wants to be able to balance with movement.
An old system I looked at: en.wikipedia.org/wiki/VirtuSphere
Not much has changed with the site since I remember last looking. Really expensive hardware ($28k)
purpose of the beads would be to allow free movement of a person that is strapped in at the waist. The wait would be mostly supported by a super structure with just a small amount applied to the bead bed.
Our goal would be to be able to detect average direction of motion and distance from the motion of the beads. This would mean we would have to have some special beads that could be tracked. One possibility could be that we imbed small magnets in beads (so plastic would be best). If we had some matrix underneath that could detect a disturbance when the different poles passed over it we could obtain a picture rotation rate. Actual rotation direction might be trickier. Perhaps if the matrix underneath was small enough compared to the magnet, direction could be determined (X/Y).
So if we actually could get rotation direction and speed from the matrix underneath we could average them together to get a sense of what the subject was doing. If they are strapped in fixed fashion then movement can correlate to using a movement joystick (forward, backward - Y, strafe left/right - X.
Balance may become an issue though as the body wants to be able to balance with movement.
An old system I looked at: en.wikipedia.org/wiki/VirtuSphere
Not much has changed with the site since I remember last looking. Really expensive hardware ($28k)