The Dozer's Almost Here!

I'm close.  So close.  Last night I managed to get functional code working.  He could roll around the arena, turn around if he hit the line around the edge, and charge if he saw a target.  Phase 1 complete.

The bot was still pretty dumb at that point though.  There was a noticeable lag in reaction time when he needed to turn from the edge, and that sometimes led to partial out of bounds.  Not good.  Also, it only sensed objects directly ahead.

So today, I refined the existing code.  I have the old stuff to fall back on, so I almost started with a total re-design.  The sonar sensor was mounted lower on the servo, since I expect my opponents to be a bit shorter then me.  I also have it sweeping around successfully, and steering the bot in the general direction of a detection, solving (kind of) the "tunnel vision" issue.

When I tested the newest code out tonight, I realized another issue.  If an object (target!) was within an inch or two of the sonar, such as when in a charge, it could loose track of them, and the bot would turn away to look again.  So I did what I swore I wouldn't do until after the competition: I added MORE SENSORS!

The large plate I was planning to use as a pusher/armor in the front still hadn't been mounted.  I have decided I'm going to add a pair of small pushbuttons behind the plate, and if either is pressed, the bot will forgo all other activities, and push as fast as it can straight ahead until it sees the line or the button is released.  I tested this with a crappy bench rig, and wrote some code.  I think it'll work pretty well.  The plate also provides the advantage of keeping the optical sensors in a consistent light.  That should help keep the readings steadier in varying conditions this weekend.

He looks angry.  Probably because I'm rushing so much of this build.  Although I have to say, I'm pretty proud of myself so far.  I'm not normally this productive in this short a time.  I guess it's good to have a goal.

Sumo Sensor Suite

Got the Ping))) finished and working tonight.  It turns out that back in the day when I received this Dorkboard, I had powered it up with a straight 9V battery like an idiot, because I was used to these kinds of boards having some kind of regulation built on them.  Since the Dorkboard takes the minimalist approach to microcontroller design, it does not, in fact, include this feature.  I noticed what I had done when I touched the top of the chip, and nearly lost my fingerprints.  Whoops.

Well, back to my point here.  It turns out that I have at least one bad digital pin on this chip now.  I know this because I fought for almost an hour like an idiot trying to figure out why the Ping wouldn't ping.  I hope that's the only issue I have...

I also got the optical sensors wired and mounted tonight.  They seem to work alright as long as I incorporate some input smoothing across the readings.  If I don't, they seem to pick up every little piece of dust and crap on the bench.  I'll try them out with some black and white paint lines later this week.

So now, it's pretty much up to coding.  Which sucks, because that happens to be my weakest area right now.  Nothing like crunch time for inspiration!

Dozer Progress

Dozer has become more like a bot then most of my previous projects.  I actually managed to fix several issues I've always had with some of my tools and materials, and that's really helped.  So the chain of design now is mainly: SketchUp + table saw + scroll saw + 800 trips to Lowe's for screws = bot.

I decided since the competition limited the bot size to a rough 6" x 6" square, I'd want to maximize my entry to take advantage of as much pushing surface as I could manage.  I ended up with a 6" circle, with the tires silightly inset to help fit.  I ended up around 6 1/4" across.  Not to shabby for roughing much of it with hot glue and some bolts.

I'm still working on the sensor code, but I do have much of the physical issues under control.  Here's a few progress pics and a short video of the lil' guy running around my carpets on a random pattern.  Mostly 'cause he's blind at that point.

 

Update: Turns out that Facebook strips YouTube videos out of blogs it imports for some reason.  If you're looking at this from there, and want to see the vid, head on over to www.hometownhacker.com and get the REAL stories!

Compact Sumobot: Dozer

TRS is planning a sumo competition soon, and I've just realized I haven't been very active in my bot building at all.  The last meet-up included some basic H-bridge and sensor instruction, so I assume that everyone is expected to use small DC motors in the bots.  The only rules I've heard however, have been that the final entries must fit in a 6"x 6" square.  After scrounging around my parts bins, I can't find any handy DC motors that would work properly as a pair.  So I'm "cheating": servos it is!  I hope it isn't decided to be unfair, the torque in these things is killer.  We shall see.

In the interest of keeping the bot as compact as I can, I'm going for an all hot glue and Velcro method.  The control board is a simple piece of Radio Shak protoboard that Rob kindly provided a few weeks back, and I'll be driving the servos and running the microcontroller from separate power sources.  I'm trying out a Dorkboard for the brains, as the Arduino servo libraries are killer, and I've been meaning to work out the interrupt issues from ADAM anyway.  Speaking of ADAM, I've officially scrapped him for parts.  Again.  Poor lil' guy... never even had a chance.

Anyway, here's a few pics of the little bit of preliminary work I've done so far.

I'm spending most of the time so far laying out all of the little bends in the wiring.  It's amazingly helpful to spend the extra few minutes on the layout so far, since I won't get so lost in the tangle I usually find myself in trying for a more "universal" design.  You can see the headers in use for left and right servos, and the small header delivering power to them up front.  Not much to look at, but sumo-bots generally aren't.

The next steps are to work on the sensor deployment and write some code.  I'm not sure what we're laying out the arena edges with yet.   That would be helpful, as if it's something I can detect with an optical sensor looking at the ground, I can keep this guy in the ring.  It's really embarrassing to turn on your creation, step back, and have it happily escape the area.  No freedom for you little bot!

UPDATE:

Been working through the wee hours.  Go insomnia.  Attempted to rig up an LM7805 regulator circuit to run the Dorkboard off the same battery as the servos.  I figured it'd be isolated enough from the current draw, but I was totally wrong.  Poor little guy can't keep up with the surges the servos pull when you move them, and the Dorkboard resets.  Bah!  The wiring was suer tight too... Oh well.  I'll just adapt what I have in there for running off a 9V cell.

Bedtime!

AVR Programming for the Fireflys

I really need to build a better harness to program these things before I solder them up.  Once I've got the LEDs soldered on, forget it.

In any case, BEHOLD!  THE SQUID OF DOOM!