View Single Post
  #8   Spotlight this post!  
Unread 08-06-2013, 14:54
otherguy's Avatar
otherguy otherguy is offline
sparkE
AKA: James
FRC #2168 (The Aluminum Falcons)
Team Role: Mentor
 
Join Date: Feb 2010
Rookie Year: 2009
Location: CT
Posts: 430
otherguy is a splendid one to beholdotherguy is a splendid one to beholdotherguy is a splendid one to beholdotherguy is a splendid one to beholdotherguy is a splendid one to beholdotherguy is a splendid one to beholdotherguy is a splendid one to behold
Re: controlling LED strips

Quote:
Originally Posted by bumblebee3339 View Post
Is that necessary to buy the same LED strip model from Adafruit to use the LPD8806 Arduino library?
I didn't actually buy my strip from adafruit. As long as the strip you buy uses the LPD8806 as the LED driver, the arduino library should work.


Quote:
Which kind of arduino should I get for this? I want it to be strong enough for image processing in the future.
We used an Arduino uno r3 clone from deal extreme, simply because it was on sale for $15. The specific one we got is no longer in stock. Most arduino models should work though. The comments in the example code referring how to wire up the strand to the arduino list quite a few:
Code:
// You can optionally use hardware SPI for faster writes, just leave out
// the data and clock pin parameters.  But this does limit use to very
// specific pins on the Arduino.  For "classic" Arduinos (Uno, Duemilanove,
// etc.), data = pin 11, clock = pin 13.  For Arduino Mega, data = pin 51,
// clock = pin 52.  For 32u4 Breakout Board+ and Teensy, data = pin B2,
// clock = pin B1.  For Leonardo, this can ONLY be done on the ICSP pins.
//LPD8806 strip = LPD8806(nLEDs);
As far as doing vision processing on an arduino goes... I wouldn't plan on it. They aren't well suited for the computations required to quickly process images. Not if you're using an IP camera at least. You could probably get something working with a camera that does some of the processing for you, like the CMUcam. Although it didn't make it on to the robot, we had very good results using openCV on the original beaglebone this year. I believe other teams have gotten rasberrypi's to work as well, although I can't really speak to how well that worked out.
__________________
http://team2168.org
Reply With Quote