Archive for the 'Vex' Category

Line-tracking robots beware

Today I volunteered at the FIRST nationals and helped set the fields up. The Vex fields, anyway. We assembled two operator control practice fields, an autonomous practice field, and a miniature, Vex-sized version of this year's FRC field. Then I was asked to tape up the lines on the regular autonomous competition field. Let me elaborate. These are the lines that competeing robots will use light sensors and such to find and follow. If they aren't just right, half the robots won't function correctly.

It disturbs me deeply that instead of hiring a team of trained experts, or at least getting someone with one of those laser-guided ultra accurate rulers, they picked me. I was armed with the following:

  • A broken tape measure. It started at the 3' mark.
  • Stretchy tape. Seriously, that stuff will easily extend an extra inch or so if you pull it out straight.
  • Thick tape. It was about 0.2' thinner then the official rules called for.

What disturbs me even more is that almost no-one else seemed to notice or care, unless a line was more then three inches off.

Ah, well. I did my best, anyway. It should be fine. But, um, ya'll do have fallback emergency routines in your code, right?


And a shout-out to Penguinator. I missed you man, sorry. Maybe tomorrow. You got my note, right?

Advertisements

All of the buffalo has been used

The Vex Challenge is over.
How did we do? Let me just say that out of about 22 teams, we came in 15th place. Pretty much everything that could go wrong, did. It was just ridiculous at times. A transmitter randomly inverted the drive-train controls. We forgot to switch the jumper chip into OperatorControl mode for one match and the robot drove into a wall. Later, we fixed one half of the drive-train just in time for a match, only to have the other half break at a critical moment about 30 seconds later.
Almost everything that could go wrong, did. With the possible exception of pretty magic smoke from the motor modules.
Oh, well.

It wasn’t all bad, though. One of our final alliance partners, team 83, was pretty awexome. They came and met with the team before the alliance selection & team captain meeting was over. How cool is that? “Tools, parts, mentors, whatever you guys need.” they said.
I don’t care who they gave the gracious professionalism award to. #83: you guys deserved it.

We also met up with one of our alliance partners from last year, #35. Almost everyone had moved on to an FRC team, but we got on with the new guys pretty well. And of course, Jeremiah the social butterfly made friends with half the people in the building. Little did he know that there was an FRC meet downstairs.

Also, there was some guy from a team I really liked last year who randomly popped up and started hanging out with us. I don’t remember what his name was or what team he was from, but I found it highly enjoyable to talk to someone not on FVC 40 who knew what I was talking about as I coded.

Sadly, this is my last year in the Vex program. Aside from the fact that I’m ageing out, my university of choice is approximately six hundred miles away. Future coders will have to get my C expertise via email and aim.

But, anyway. We had a great team this year. FVC 40 will be a great team next year. Stampede on!

On the state of autnomous mode

The robot does not do anything in autonomous mode. With mere hours to go, I suppose this should worry me, but I’ve been desensitized to it by the fact it hasn’t done much of anything all week. Now, I can only watch in horror and eat taco chips as 7:30 AM, March 16th draws slowly closer.

And then, disaster struck.

Ok, so I finally got around to attempting to drive the robot manually. It didn’t work.

All the sensors and motors work fine with the online test program, wireless & tethered work the same, and I’ve tried three different transmitters. The transmitter is in mode 23 and it seems to make no difference if a jumper is in port 15. Yes, I’ve got a jumper in interrupt 6.

All that’s in the operator control function is a statement assigning tank mode to motors 2 and 3. In the process of debugging, I’ve cleared away everything else.

The scrimmage is in less then eight hours. Joy. Oh, well. Maybe I’ll find someone who can help there.

Vex Update 3ish

I’ve become the more-or-less unofficial electrician for the team. No one else did anything with the electronics at all, so I’ve figured out what needs to go where and sketched out a little graph of it.

I’m now using this graph as a refrence for my more-or-less official coding position on the team. As of tonight, the autonomous code can:

  • Follow a line (sorta)
  • Make a leap of faith from one line to another (untested)
  • Turn a given angle (With acuracy of within about 20 degrees. Needs fixing.)
  • Sense a wall about a foot in front of the ‘bot
  • …and stop before hitting said wall

That last one might sound obvious, but you’d be surprised… :-p

What’s kinda scary is that the robot is pretty much built, but we’ve never had everything hooked up and running at once! The reason is, all our electronics are mounted around the parimiter of the robot; we have a sort of basket in the middle to store balls. This means all the cords have to go around the side of the basket, and the ones connected to the motors and sensors on the back of the robot are not long enough. We have some extenders on order, hopefuly they’ll get here before the Scrimmage on the 11th.

The Eleventh. That’s not even funny. The Eleventh. That’s just rediculous

An excerpt from the bill of materials…

Zip ties – 66

Wow. Seriously. Keep in mind, this robot is only about the size of a breadbox. And no, we have not replaced all the screws with zip ties – yet.

Do you like donating money to worthy causes? Does your company make little black zip-ties? Contact us today for information on becoming an official FVC-40 sponsor!

And work on the robot plods onward

Slowly. We’ve worked on it all evening. Our ball basket it much lighter, we may or may not have a way of dumping balls now, and we have consumed massive amounts of junk food.

Our main problem now is that no one feels like building said ball dumping mechanism because to test it out, we’ll have to integrate it with the chassis. That’s the problem with having a small, light robot. It’s not very modular.

Our other problem involves the beater-bar. It’s great at grabbing balls, but it’s getting in the way of most of our ball-dumping ideas.

Jeremiah and I have been practicing our driving skills by playing Wild Metal Country. With the proper control layout, playing the game is very much like driving a Vex robot equipped with four wheel drive. I dunno if I’m gonna drive or not, but I figured Jer needed to be took down a few notches.
He drives OK, but I’m concerned about his strategy & game theory abilities. If confronted with a catapult rolling bombs off the side of a cliff, he’ll try to drive right up the cliff. Never mind that a 20 second detour would put him on level ground with his nemesis and the rolling bombs destroy him every time.
In multiplayer, even if I take his shield down to almost nothing, he’ll attack head on, driving right into my barrage of missiles. Not to mention when I was afk, he hunted down my non-responsive bot and blew it to bits. Nice gracious professionalism, eh?
Like I said, I’m not too sure about him. I dunno if I wanna drive again this year, but I’m hesitant about Jer representing our team’s robot and strategies on the field.

easyC 2.0 has been obtained, installed, patched and registered. I’m too tired to explain exactly what happened, but you can figure it out from this ChiefDelphi forum thread.

Someone said something funny earlier this evening during the build session, but I don’t remember what it was. My tired mind has begun to filter out what people say. That’s what I get for staying up till 2:30 AM working on this thing. Meh.
I’m gonna go to bed.

*yawn*


RSS Status

  • An error has occurred; the feed is probably down. Try again later.

Del.icio.us

Creative Commons License
This stuff is licensed under a Creative Commons License.