Forgot your password?
typodupeerror
Robotics Hardware Technology

Harvard Licenses Technology For Tiny Swarming Robot 38

Posted by samzenpus
from the own-the-swarm dept.
Zothecula writes "Do you think that you'll never be able to afford a robot of your own that isn't a toy? Well, if you can get Swiss robot-maker K-Team Corporation to sell you one, chances are you can easily afford a Kilobot — perhaps even a whole bunch of them. Designed and first built by Harvard University's Self-Organizing Systems Research Group, the three-legged robots aren't much larger than the 3.4-volt button cell batteries that power them, and move by vibrating across smooth, flat surfaces. They were created to study robotic swarming behavior, with the intention that tens, hundreds or even thousands of them could be used simultaneously in one experiment. Harvard has just announced that it has licensed the Kilobot technology to K-Team, which will commercially manufacture the robots so that other groups and institutions can purchase them for their own research."
This discussion has been archived. No new comments can be posted.

Harvard Licenses Technology For Tiny Swarming Robot

Comments Filter:
  • by Jawnn (445279) on Sunday November 27, 2011 @11:56AM (#38182140)
    ...to build copies of themselves. What could possibly go wrong?
  • Reliability (Score:4, Insightful)

    by Hentes (2461350) on Sunday November 27, 2011 @12:07PM (#38182196)

    Moving using vibration could be hard on those tiny legs. What is the lifetime if these things?

    • Why use these when you can just simulate them? Designing/training the AI would much more interesting than seeing if it can be carried out by this particular design.
      • by Hentes (2461350)

        I guess they do use simulations to refine the original design as it's way faster, but eventually every algorithm has to be tested IRL. Simulations are imperfect.

  • by sconeu (64226) on Sunday November 27, 2011 @12:12PM (#38182232) Homepage Journal

    Just what we need... Killbots.

    • by Anonymous Coward

      Costello : Finally, a KillAbott.

    • "Ladies and gentlemen, my Killbot features Lotus Notes and a machine gun. It is the finest available."

    • Not a problem, we can outsmart them:
      "Killbots have a pre-set kill-limit. Knowing their weakness, I sent wave after wave of my own men at them, 'till they reached their limit and shut down"
  • by HizookRobotics (1722346) on Sunday November 27, 2011 @12:53PM (#38182424) Homepage

    There's a long history of home-made steerable vibrobots [hizook.com]. You can probably make one of these from parts readily available in your junkbox.

    A quick tangent: I've seen these in person. They're pretty cool, but I'm not sure what "technology" Harvard is licensing. Perhaps just the PCB design and code?

    • by Hentes (2461350)

      It's not just the motion but also the communication using reflected lights.

      • I understand that they can communicate using reflective infrared light. My TV remote can do the same thing to reach a non-line of sight receiver. I believe they also use the reflected IR light for rudimentary distance sensing -- much like the Sharp IR sensor modules. What I'm trying to say: the hardware aspects of this project are fairly well-established -- as far as I can tell, there are no new "hardware" technologies (but I am a fan of readily-available low-cost robots). The Harvard group's big contri
    • by mako1138 (837520)

      Well, the things are ~$1200 for 10, so someone's making money, I bet.

  • The coin! Looks foreign to me. If that's a robot the coin must be a huge. I want one.
    Hope they ship to The Netherlands!

    • The coin! Looks foreign to me.

      That's a US quarter dollar coin. Wikipedia says they're 24.26 mm in diameter. FYI.

  • by MichaelCrawford (610140) on Sunday November 27, 2011 @01:24PM (#38182626) Homepage Journal

    -s?

    Isn't the whole point of academic research the free and open sharing of knowledge? But universities obtain patents then sell exclusive licenses to them, despite the research generally being done at taxpayer expense.

    • by Missing.Matter (1845576) on Sunday November 27, 2011 @02:02PM (#38182850)

      Here is the IP policiy of the NSF: http://www.nsf.gov/pubs/manuals/gpm05_131/gpm7.jsp#730 [nsf.gov]

      This project in particular was funded by the NSF and the Wyss Institute, which looks like some sort of incubator funded by Harvard. The gist of the NSF policty is that the grantee retains all rights to the invention/patents/copyrights of the research. This makes sense given the NSF's mission statement: "To promote the progress of science; to advance the national health, prosperity, and welfare; to secure the national defense...."

      • by Anonymous Coward

        The robot design is also open source under a creative commons license that lets you make your own ...

    • Isn't the whole point of academic research the free and open sharing of knowledge? But universities obtain patents then sell exclusive licenses to them, despite the research generally being done at taxpayer expense.

      The licensing can be done is a way that favors the taxpayers as well. For example the University of California is quite aggressive about patenting discoveries and licensing the patents. The terms of the licensing is far more favorable if you are a startup or small rather than a large established corporation, if you are local, etc. I'm not sure but there may also be consideration for companies that support research at the university or employ its students.

      Half the fees collected go to the UC system in gen

  • by Smurf (7981) on Sunday November 27, 2011 @01:46PM (#38182758)

    TFS says:

    Do you think that you'll never be able to afford a robot of your own that isn't a toy? Well, if you can get [...]

    Cool. Clearly these "three-legged robots (that) aren't much larger than the 3.4-volt button cell batteries that power them, and move by vibrating across smooth, flat surfaces" are very useful if you are "studying robotic swarming behavior" and I bet people can come with several other fields where these mini-robots could be of practical use.

    But then, most of us would be able to come up with several experiments which would make practical use of any of the many robots that are currently sold as toys. And some of those experiments would without any doubt be commendable research projects.

    Unfortunately most of us do not work in those areas of research where the Kilobots (or any other commercially available, reasonably priced robot) are applicable. For 99.9% of the people, Kilobots are nothing more than FREAKING TOYS!!

  • (/me ducks)

  • Oh well, we can always build more Kilobots

Porsche: there simply is no substitute. -- Risky Business

Working...