Slashdot Mirror

← Back to Stories (view on slashdot.org)

GUIs for Robots

Posted by michael on from the click-drag-boom dept.

kabir writes "OpenGL.org has a link to a fun article over at the Stanford Aerospace Robotics Laboratory. It seems an OpenGL-based GUI has been developed to enable the operation of multiple complex field robots by a single operator. The interaction mechanism was inspired by interface techniques refined in the Real-Time Strategy (RTS) genre of video games. Fifty years from now I expect all wars to be fought by giant robots controlled by teenagers."

3 of 156 comments (clear)

  1. Re:Not sure how long this will stay up, so... by loply · · Score: 2, Interesting

    I did the same thing for a little 3D platform I made (which I want(ed) to mount an airgun on).

    It was parport controlled and I used OpenGL to draw a crappy picture of it onscreen in a Qt application =D

    --
    loply.com
  2. Voice activated by dpavlenkov · · Score: 1, Interesting

    Why voice? put a retina tracker on your eye, blink to lock on target, blink again to unlock, and blink twice to destroy. You can use your voice to do other useless things in the meanwhile.

  3. Questions to developers... by josh+crawley · · Score: 4, Interesting

    Article said:
    ---"However, the nature of field robotics requires some significant differences in the implementation of the RTS interface method. For instance, there is no single source of accurate global information -- each robot can only provide relative data that has to be fused together. In addition, the tasks that each robot can perform change dynamically and this information must be reflected in the choices presented by the GUI to the operator"

    Well, umm. Yes there is a single source of global information. It's called GPS. I believe, using the correct civillian gear, you can get centimeter-accurate location points. Also, political/geographic maps are available from many locations from different governments.

    The best way I see that this can be used is that you create a neural network between the robots by way of bluetooth. The robots could share GPS location data in realtime. Make sure that they can pass on data to out-of-range units.

    Once you have location links, you can piece screenshots (by using angles of the GPS coords). Personally, I'd craft chips designed specifically for piecing the screenshots to a 3d locale. No sense having a generic chip (x86 or whatnot) doing that. And NO beowulf clusters :-)