Don't ad-block us - support your favorite websites. We have safe, unobstrusive, robotics related ads that you actually want to see - see here for more.
0 Members and 1 Guest are viewing this topic.
I mean is there a way to distinguish at least four different infrared leds by (infrared) camera?
and detect them via an IR detector rather than a camera.
Ooh, ya. That'd make it challenging. There's probably a way, but complicated and finicky...
But if somebody know a way how to do it with infra leds (or other type of led) I would be very grateful.Btw. on top of every robot there are three leds in a row. In same distance on each robot. So detection is not done by positioning leds.
I just need to use IR/Uv paint to create the QR code and it should work. Then I have more than enough combination and detecting QR via image recognition is more reliable than using led (I think).
Soeren,good thing is that my robot's size is around 10cmx10cm ant top is flat co QR can cover most of it. There are supposed to be 5 - 10 object on the field so I don't need so many combinations. Then QR can be low res. In this situation it should be simple task to capture and decode it with camera.
Hi spooneythegoon,and how would I determine position of the robot on field? Btw. robot are controlled by PC with BT.
In that case, you might wanna consider using bar codes, code 2/5i
Do you just need to track it?
Dut what I was thinking about was actually using my own version of QR. Let say with resolution 3x3. If we remove two fields to identify front and back side we still have 7 fields that holds 2**7 = 128 combinations. And since we are programmers it should not be a problem to implement this pseudo QR code decoding algorithm.What do you think?