Defying gravity for new user interfaces

[Jinha Lee] at the MIT Media Lab created a new interface allowing new ways to modify and play with 3D objects. It’s called ZeroN, and it’s nothing short of a futuristic device straight out of Star Trek.

ZeroN is simply a platform for levitating a small spherical permanent magnet in 3D space. It does this by mounting a hall effect sensor on an electromagnet. The hall sensor measure the strength of the magnetic field of the ball every few milliseconds and keeps the spherical magnet levitated. To move the object in 3D space, a few webcams track the ball over the platform and tell the electromagnet to move on a CNC-like x y table.

[Jinha] showed off a lot of cool stuff that is possible with the ZeroN; ping-pong is by far the coolest implementation, but it’s also possible to use the magnetic sphere to demonstrate n-body gravity or as a camera flying around a digital scene.

It’s a really amazing piece of work with an exceptional demo video. You can check that video out after the break. Thanks, [ferdinand] for sending this one in.

Comments

  1. Daniel says:

    I’d like to know how much noise it makes during use. And I would like to see it handle two balls.

  2. Max says:

    They should’ve used repulsion via eddy current instead of attraction. this would make the effect even cooler by levetating an aluminum ball in mid-air, without the big base above the whole assembly and position information is there via webcams…

    • Drake says:

      Wouldn’t that require a /Moving/ magnetic field?

    • David says:

      Sadly, eddy currents like that only serve to resist and slow down motion that would otherwise happen. You can effectively slow gravity, but not stop it (levitate); and certainly not make them move in a different direction (e.g. up, against gravity).

      • Zak says:

        Eddy currents can provide lift. If you take an aluminium tube and have a magnet fixed to a rod inside the tube, lifting the magnet from the tube provide plenty of force, and the tube lifts with it. It does slide down slowly of course, but gravity is overcome.

        How this would be applied to this concept I don’t really know.

    • Max says:

      It can provide lift very well. If you have an AC powered coil with enough current, nearly everything conductive will pop-out (that’s why induction-cooking-plates don’t work with non magnetic pots)…

    • Dbear says:

      eddy currents also would generate heat. Touch a pan on a induction stove sometime.

  3. IJ Dee-Vo says:

    Nice, add a pattern to the sphere and its orientation can be tracted as well.

    • Brian Neeley says:

      I was just thinking the same thing. The problem is, although the ball is spherical, it is still a magnet, and would have a distinct UP and a DOWN. I’m not sure if you would have that problem if using the setup Max suggests (above). My thoughts would be a something along the lines of a metalic sphere, surrounded by a plastic shell, covered in little black dots (like some optical mouse balls), and perhaps a differently colored dot to denote FRONT. There may need to be a small bit of fluid between the sphere and shell for lubrication, but other than that, I don’t know how to improve on this concept.

      Awesome concept in any case.

  4. wardy says:

    The jerky motion of that steel ball makes me think there’s a pretty hefty magnetic field at work here. Does this therefore pose a potential hazard to people with medical augmentation (pacemakers / metal pins / cochlear implants etc)?

    • b1r6m4n says:

      I thought medical metal inside bodies was never ferromagnetic??

    • ferdinand says:

      i think a peasmaker can be a probleem i no that pepole whit one not can use a induction cock plate

    • rasz says:

      its jerky because the ball is wobbling on the edge of magnetic force while being pulled by the XY arm

      • And00m says:

        Believe it will also have to do with the latency/refresh rate of the images from the cameras as they process through the PC and give a physical product.

    • Chris says:

      By requirement, most of the field is concentrated in the vertical axis, and will only be encountered through the small area intended for interaction.

      Any implants in body parts intended to be placed in that area shouldn’t be ferromagnetic.

      Electronic implants with copper wiring can generate induced currents in response to rapidly changing magnetic fields, like the cook plate example which is powered by AC. The coil in this device is PWM’ed, but they don’t say at what rate. If fast enough, the field could be nearly constant, and the swinging of the ball merely pendulum effect within the non-rigid confine of the field.

      I suppose there could still be induced currents from someone with an implant walking through magnetic field lines leaked horizontally, though. It’s a good question and probably worthy of some tests if this is to be publicly exhibited.

  5. Kyoorius says:
  6. Dra says:

    That would be awesome as a input device for 3d modeling and the like. Move ball to point in relation to point on mesh, and physically manipulate the point.

  7. conundrum says:

    Intriguing, thanks for posting.

    I expect that using an induction sensor in the ball would also work, by simple inverse square law.
    aka power drops off with square of distance.
    Superimpose a high frequency signal onto the drive signals, and then use a receiver inside the ball to measure each and broadcast over a 433.92 MHz link.

  8. Gert says:

    Imagine implementing this in a sience museum but with the solar system.

    • Kris Lee says:

      The problem is that the ball here is moved around mechanically.

      You can see this around 2:30, specially starting from 2:35 in the video.

      You can move only on ball around like that.

      This is also the reason why the movement is shaky.

      The magnetic force between the ball and moving head is more like a string and the ball stays in place or moves forward more than desired because of inertia.

      Doing your idea with this system is not impossible but the guidance rails should get much, much thinner so it would be possible to stack them.

  9. Pascal says:

    I want to go to the MIT 😀

  10. blujay42 says:

    jesus why does he keep the finger

  11. Kris Lee says:

    It looks like that this globe http://www.youtube.com/watch?v=uW7Rd6Nn2-Y neads a small upgrade.

  12. Kris Lee says:

    And it looks like it is based on Fab In A Box platform: http://mtm.cba.mit.edu/fabinabox/blog.html

  13. Ant says:

    Why not use some sort of delta robot setup? I bet that would provide a more fluid form of motion.

  14. sutapest says:

    why not use a smaller weight?

  15. mic says:

    Now add hand tracking with a kinect like system. That woud trip out your friends I bet.

  16. Sobiguy says:

    I think their usage ides’s are awesome, But not until you can get a usable area more than 5 or 6 inches from the top. Notice they manually moved the ball wear they pleased but the magnets ability to control the ball never exceeded more than say 5 inches. So the rest of that area below it is pretty much wasted…

  17. asz says:

    Neat, but I like the version that uses three light jets of air to hold a pin-pong-ball-ish sphere in three dimensional space. It used IR LEDs instead of an expenisve kinect and the sphere could be 15″ away.

    Looking for link…

Speak Your Mind

*

Related Hacks in peripherals hacks

  • RSS Feed Testing
  • Hacking the Kingston Wi-Drive
  • Putting a software defined radio on a mac
  • Defying gravity for new user interfaces
  • [Matt's] bubble display updated