Kit's Blog

Creative Writing is excellent

     So tomorrow I'm going to post some stories I've written while in Creative writing. For now, all you need know is that I'm trying to do homework, and I'm totally not posting on my blog in order to avoid it... especially not at 12:23 in the morning... nope... not posting.

Oh, and in case you were wondering how techno-babbly I can sound check out this email I sent out to the robotics programming team (a quick side note should tell you that THEY didn't understand it either):

////////////////////////////////subject: A cookie for anyone who can integrate 200 times a second
////////////////////////////////Body:
Dear everybody on the programming crew,

So with build season upon us it looks like only Matt, Brian, and I care. Oh well, such is life. Anyway, Just to toss a challenge at you that you'll never complete, who can make this happen?

I need to figure out how to read the data the accelerometer gives us. Right now we can get the actual data, but only from one of the ports. Anybody who can get more than one of the ports working gets a cookie (yes, I'll be bringing in cookies, a real one).

Secondly, I need to figure how to integrate the whole thing. This is just basically multiplying it by the amount of time that has passed. But you also need to make it into workable units. Here's the killer, the accelerometer gives data as 290milliVolts per 1 g of acceleration. That's 290milliVolts per 9.8m/s/s. The ADC.c program spits out wildly different values depending on the number of samples you take (look in ADC.h). I need someone to look into what exactly "u"volts are. I think he means *10^-4 but I'm not certain. I'm not asking for the wildly complicated integration, I just want to see who can figure out the conversion factor from what Kevin gives us to an actual unit of measurement.

Thirdly (and probably lastly), I'm looking for anyone who can figure out how to have that whole system integrate at 200 times a second within the ADC program. Okay, no one can do that but me... nevermind.

--Kit

Attached files:
frc_gyro.zip -- This is the code for this year's gyros. You'll want to mess with this MOST of the time. If you think you've got better ideas then go for frc_adc.zip on kevin's website. Most of this code won't be useful until we get to the yaw-rate-gyro, but the bias function will be very useful.

frc_accelerometer.zip -- This code has the conversion factor for the 2005 accelerometer. Don't know if it will help, but it's there if you need it.

Posted by Kickmyassman at January 26, 2007 12:30 AM