Log in

View Full Version : Additional Processors

chris31
21-09-2006, 18:26
Last year I was team leader and programming didnt get much of my time due to leading. But this year I want to do more programming. Im curiuos as to how many of you use extra processors to take load off the RC. If you do tell me a little about what you used.
AdamHeard
21-09-2006, 18:48
Several teams have used the gumstix brand processor before.

200 and 400mhz models under $200
Qbranch
21-09-2006, 19:53
Question: what on earth could you be doing that you need a spare processor? except for vision of course, but that's handled by the fujitsu in the CMUcam....

-Q
Dave Flowerday
21-09-2006, 20:04
Question: what on earth could you be doing that you need a spare processor?
Plenty. We've used (and needed) a secondary processor every year since 2002.
AdamHeard
21-09-2006, 20:07
Plenty. We've used (and needed) a secondary processor every year since 2002.

Could you elaborate as to why and how it helped?
Dave Flowerday
21-09-2006, 20:44
Could you elaborate as to why and how it helped?
You can check out some of the threads where we've discussed some of the things we've done:

StangSense: http://www.chiefdelphi.com/forums/showthread.php?t=3302
StangPS: http://www.chiefdelphi.com/forums/showthread.php?t=20273

We used enhanced versions of StangPS in 2004 and 2005. Also in 2005 we added camera support to our custom circuit and used that information combined with the position info to approach the tetra correctly (our robot required that we drive parallel to one side of the tetra no further than 2 inches away) and this year we implemented an auto-tracking turret. 2006 is probably the first year where our software could have run entirely on the RC, but having the separation of tasks makes it more manageable to implement and debug.
Mike
21-09-2006, 21:52
StangPS... *drool*

I've been working on something a little bit similar to StangPS that has required an additional processor. I, personally, don't have the time to develop something advanced enough to need a Gumstix (yet) so my projects use the Atmel AVR ATMega series microchip.

I'd recommend checking out the STK500 development board. A friend of a friend described it as the "swiss army knife of avr development" and I tend to agree with him.
chris31
22-09-2006, 10:13
You can check out some of the threads where we've discussed some of the things we've done:

StangSense: http://www.chiefdelphi.com/forums/showthread.php?t=3302
StangPS: http://www.chiefdelphi.com/forums/showthread.php?t=20273

We used enhanced versions of StangPS in 2004 and 2005. Also in 2005 we added camera support to our custom circuit and used that information combined with the position info to approach the tetra correctly (our robot required that we drive parallel to one side of the tetra no further than 2 inches away) and this year we implemented an auto-tracking turret. 2006 is probably the first year where our software could have run entirely on the RC, but having the separation of tasks makes it more manageable to implement and debug.

Pretty impressive work.
RoboMadi
22-09-2006, 10:25
Another processor can be useful for a lot of functions. I know couple of teams use it for navigation/positioning system. I've seen other use it for sensors. As of now, I know Team 612 is working on a voice recognition processor. That would be cool though; instead of joysticks, drivers will be using headsets. (Crazy F.I.R.S.T. ideas :p )
chris31
22-09-2006, 10:34
Another processor can be useful for a lot of functions. I know couple of teams use it for navigation/positioning system. I've seen other use it for sensors. As of now, I know Team 612 is working on a voice recognition processor. That would be cool though; instead of joysticks, drivers will be using headsets. (Crazy F.I.R.S.T. ideas :p )

Sounds cool, would take time and also making sure that it works in the noisy conditions of competition is important. Maybe its just me, but i can hit buttons faster than i can say commands. And then when the command isnt being recognized your left there useless, unless you have button backups.
RoboMadi
22-09-2006, 11:06
Sounds cool, would take time and also making sure that it works in the noisy conditions of competition is important. Maybe its just me, but i can hit buttons faster than i can say commands. And then when the command isnt being recognized your left there useless, unless you have button backups.

As i mentioned its a crazy idea (I was just joshing). I agree with you, operating your robot with this kind of system would not be easy.
And just for clarification, Team 612 is working on a Voice-Activated Wheel Chair for a disable kid; and thats why they're using that particular processor.
chris31
22-09-2006, 13:35
As i mentioned its a crazy idea (I was just joshing). I agree with you, operating your robot with this kind of system would not be easy.
And just for clarification, Team 612 is working on a Voice-Activated Wheel Chair for a disable kid; and thats why they're using that particular processor.

Oh, a very good cause.

ats why they're using that particular processor.

You never said what processor
RoboMadi
22-09-2006, 23:07
I know Team 612 is working on a voice recognition processor.

Heres (http://www.mikroe.com/en/tools/easyvrstamp/) all the details