Subject:
|
Re: 3T
|
Newsgroups:
|
lugnet.org.us.laflrc
|
Date:
|
Sat, 31 Dec 2005 15:53:46 GMT
|
Viewed:
|
1450 times
|
| |
|
|
In lugnet.org.us.laflrc, Brian Davis wrote:
| |
In lugnet.org.us.laflrc, John Brost wrote:
| |
I’m finding it near impossible to use a light sensor and
tell the difference between a black block and no block...
I guess I’m going to have to re-think my entire block sensing
solution.
|
I finally ran into the same problem at high speed last week. Rats. With or
without an accessory light, the difference are just too small to deal with...
or at least trust. So I took off my nice, compact light sensor, and had to
invent a mechanical one. Oh well, I learned, and it seems to work. Still not
100%, but it’ll get there. And I could still potentially determine block color
if I really wanted to.
|
Hmm..how strange...my T3 robot uses a light sensor to find the blocks and I am
having no trouble at all distinguishing all the information I am after. I think
I am getting a light delta of around 20 or 30. That seems to be plenty.
But, I am not telling you how I did it because I am pretty convinced it gives me
an advantage right now :) sorry.
Wish I could be there Monday to show it off.
Oh - and just in case anyone needs a target to shoot for...I haven’t actually
timed my robot, but I think an average play from when I receive a signal to when
I give a signal is about 4 seconds.
I have put WAY too much time in this robot :) But loved every minute of it.
I had to re-learn C language and truly figure out Brick OS to do it. I also
found (second to the sensor mechanism) the most difficult thing to be the
routing and controlling of wires because there are so many of them :)
My robot has 3 light sensors, 4 touch sensors and 3 motors (and a partridge in a
pear tree). I tested portability this week on a 3 hour drive to a funeral and
back. The robot made the journey with no hitches (though I didn’t do a speed
bump test for Steve).
I even built in some fault isolation and recovery in the software...pretty
scary.
Anyway, if you can’t tell by what I have written, I am excited about having this
one under my belt. I am looking forward to seeing your alternative solutions
(especially for the sensors).
Good luck everyone.
Bryan
|
|
Message has 2 Replies:
 | | Re: 3T
|
| (...) Why would you think that's strange? That's the normal difference you'd get between black and white. While John and Brian are pointing their light sensor at the cube/square to check for one of three states (black cube/white cube/no cube) you're (...) (19 years ago, 31-Dec-05, to lugnet.org.us.laflrc)
| | | Re: 3T
|
| (...) Well, sort of. You seem to be using a light sensor to read a mechanical system, not observe the blocks themselves <grin>. And, as my post implied, I ended up with a mechanical system. It's just it doesn't allow all the fun stuff I wanted it to (...) (19 years ago, 1-Jan-06, to lugnet.org.us.laflrc, FTX)
|
Message is in Reply To:
| | Re: 3T
|
| (...) I finally ran into the same problem at high speed last week. Rats. With or without an accessory light, the difference are just too small to deal with... or at least trust. So I took off my nice, compact light sensor, and had to invent a (...) (19 years ago, 23-Dec-05, to lugnet.org.us.laflrc)
|
19 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
