by Stark Raving Mad
We made a complicated bot. We pushed the limits of every rule. Our tether was a flimsy chain looped around an axle at both ends to cover that "must be connect by lego" rule. We shoved our extension arms in the cubic foot to get impounding approval, only to have our size contested AGAIN on contest day. We put our swatch velcro in an approved place the night before, and then during calibration we were told we had to move it to another spot. All this caused our first "delay of game". Whatever the heck you wanna say about it, however you feel like ridiculing it, we chose to ignore the ole "Keep It Simple Stupid" and decided to give ourselves a challenge. And we pulled it off. We just weren't able to show everyone that we did.
We were barely in class this week. We kinda missed part sorting just because we have 13-hour flights from Hawaii, which means leaving home on or before New Year's Day. We both barely made it to class on time on the first day. The rest of the week we spent maybe 30 minutes total in workshops and quite possibly over 12 hours of Halo2. We got through the first two assignments with a dummy bot pretty quickly.
So snowboarding began and Chris had to leave two nights a week. Curtis came on board, but he had to work from 1 to 5 everyday. We began formulating our strategy, keeping in mind we had two MechE's and only 1 programmer. The third checkoff took awhile for a few reasons:
Thankfully, we caught the problem, and when it worked, our dummy bot found that swatch like a heat-seeking missile. One guy who was there all day came up to us and said, "You guys are BADASS."
- our Reckless Programmer stubbornly wanted to implement a continuous drive system, unlike all the stop-and-go driving of 90% of the teams
- the turn function he implemented was slowing down the motors indefinitely
The 4th checkpoint was battery assembly and contest strategy. See our strategy here.
This week was full of... no, not all work. It was BIRTHDAY MADNESS and DRUNKEN REVELRY. Yesssssss.
But seriously... We got the initial orientation mechanism to work. We were using a servo geared to a wheel sitting face down to turn the bot around an axis through the wheel. Think of it like a rotating platform. We had to gear the servo down to get 270 degrees of rotation (the minimum to account for all orientations). And because we geared the servo down, we weren't getting enough torque, axles were moving, gears were slipping. This really sucked. Until we got a continuous servo. Rated at the same torque as the standard Futaba, we would still have more torque because we didn't need to gear it down to achieve 270 degrees of rotation. This solved our orientation problem when we tested it at home.
We got the Halfback's line following to work at home too, using electric tape on the kitchen floors. We even got the dismount off the stationary bot to work. The arm extensions were working nicely as well. The pieces were working quite nicely individually. All that remained was to put it all together and test it out in lab. But there were two critical things that we didn't think off until the final day.
Thing No. 1: The Tether
At about 7pm the night before impounding we discovered our original idea for a tether wouldn't work. Random pieces connected by pegs coiled around a pole. Firstly, the weight of the wires would sag and drag. And second, the dismount would pull too hard on the pole, yanking it out or not uncoiling well. We sat there for a long time, considering alternatives like scrapping one of our two bots for parts and totally revamping our strategy. We were pretty disheartened at this point. Curtis, clear-headed as ever, told us, "Ne boisia, boys. Shiver not. We shouldn't just throw away old ideas as soon as we run into an obstacle. We haven't had enough time to think of solutions. There is a way."
And with that, he left to play IM basketball while we tried to think of solutions. Thanks Curtis.
A troop of 4 friends came by and delivered dinner to Chris and Ricky. It was great to finally eat and see some faces besides the ones in lab. We told them our problem and within half an hour we had our solution. We used the lego chain (the tiny black links) and taped it to our wires in segments of about 4 inches. This allowed folding of the tether in such a way that the dismount would unfold the tether without snagging. We credit our tether to these fantastic friends. Thanks Terence, Hannah, Jon and Jon. We love you.
Before we mention Thing No. 2, we should mention that our sensors were acting up and set us back about 2 or 3 hours. The stationary bot was leaning, which put some sensors at the wrong height. Also some of the solder connections were breaking despite the hot glue reinforcement, so we had to redo those. This was a hard problem to identify on 2 hours of sleep at 3am on an empty stomach.
After solving this setback, things were beginning to come together. We had maybe 3 consecutive successful runs on the test table near the elevators. We got our qualification checkoff in lab. There was fear in the eyes of our spectators.
We were set, except for one thing...
Thing No. 2: Friction
The boards were repainted on contest day. Fresh, untainted by the scuffling of 50 robots repeatedly grinding down the wood, untarnished by the dust of lab and the skin flakes of geeks who haven't showered in several days. This was to be our downfall. The arms had heavy friction with the board on contest day, preventing our pivot servo from providing enough torque to turn the bot. Alas, our bot moved 60 degrees, and only once. We were ridiculed, shamed, disappointed. What's worse is that a crowd gathered around our bot on its first run. And we let them down.
What ball-gathering, one-way gated, differential-drive, generic bot would garner such attention? NONE. We had the support of the Ball Sorters, the other Extendy Arm group, and Hinjew, just to name a few. Those guys had cool ideas, they wanted to be different too. We had to stick with our complex, unique, assembly-required design for those guys. To give them the glimmer of hope that something different could work. That K.I.S.S. was for wimps. That Legos could do crazy things they were never meant to do. That our bot could be Ghetto Fabulous.