Winter Testing

Trying my hand at arts-y robot photos.
Many thanks to Royal Robotics for hosting this practice field!

I may have mentioned before (I can't remember everything I write!) but I am a huge fan of Formula 1 and Formula E. The similarities to FRC are quite striking. Million dollar deals, hundred-million strong fan base, cars... These are all things that make Formula 1 what it is today. Anyway - this time of the year especially, the parallels are exciting to see. Waiting for reveals, testing videos, shakedowns, reliability and performance tuning - not just talking F1 with that list, we're doing the exact same things in FRC. (Seriously - teams! More reveal videos! Do it! Also, I'm told Baby Yoda is watching. Don't disappoint Baby Yoda.)

This time certainly feels like winter testing. Our drivers are getting their first hands on the controllers. Our full systems shakedowns, mechanical tweaking and tuning, software tweaking and tuning are all full-steam ahead. (Was that a 2017 joke?). We're looking to prove out our reliability, while starting to push our performance.

Celebrations!

As we are now in week 7 (Starting 8? Who can keep track anymore...), Tuesday held a special day - the now-defunct Bag Day. CyberKnights held a small memorial service, a minute of silence, then ate some cake.

Not a lie.
Note the... dead... eye.... :-D

Okay, we didn't have a moment's silence, but we did reflect that Bag Day came and went - and we still have work to do. (Fortunately no longer bound by 6-hour windows and withholding limits!) Our new head coach brought along a modified version of an old tradition, and we recognized our seniors - allowing them to blow out the cake candles. (Unless they had a cold. No blowing across everyone's cake. No robo-plague!) Once again, it's nice to take a moment, celebrate the small things, recognize team members, and just take in a breathe of air.

Another small victory this week was our climber CAD - constantly pushed and iterated and modified as other things changed - passed final CAD review this week and started manufacturing! Go Team! Woo! (Subtitle: And there was much rejoicing). 

Can't climb just yet, but we're getting there!

The Good, The Bad, and The Ugly

Let's start with the good. No sense putting an ordered list and then not following it, eh?

The Good

Om nom nom.

This robot has good bones. (Apparently some kids have never heard this expression before. "Has good bones" means that the core of the robot "is lit". Or, stated another way, the core execution is "pig". Pog? Pag? Lit.)  This is going to be a good. (Literally as I wrote this, software just shot 4/5 Inner Port. Derp. Er- Dope.) Right - as stated earlier - so many teams come up with similar priority lists, similar methods of execution, similar mechanisms. It's the details that start to separate the good from the great. Right now, we're good, but we have head-room to develop and become great.

Self-team-promotion - great job designers, machinists, assemblers, and software. As we continue to iterate, test, and develop (likely, up to and including load-in day at Glacier Peak), this robot will just get better and better. With climber also passing CAD final review, we <should> be able to capture more points and assist in the Generator Switch RP. Overall, the whole team is looking pretty excited this week.

Feel good, perform well.

Formula 1 reference: Lewis Hamilton says Formula 1 is his relaxing fun time away from his other pursuits in fashion, music, and business. If this is his "unwind" time and he's performing at such a high level, how can he possibly be stopped?"

The Bad

I've noticed when I chat with folks, both in the lab, on Slack, and elsewhere, I can sometimes get pessimistic and have to remind myself that we have a lot of positives. Fortunately, with this section, I can describe everything bad that I want! Bwahaha!

We're finding the bones are good, we're getting our final sub-systems manufactured, and we're performing our dance of mechanical and software tweaks and tuning as we find issues. The problem is that these really should be week 5 tasks, and we're in week 7. Number 1, absolute, apex, peak, critical, crucial requirements for a successful robot are reliability. If we don't have hundreds of hours of practice time logged in the lab and the practice field, we haven't found enough of our failure points. So we will find them on the competition field. That's not where you want to find out that you can snap a pulley, or a part rotation pulls on a wire, or the pneumatics can't operate under certain load conditions. Find these things out as early as possible.

The math worked... for a spherical intake in a vacuum!

A lot of things can lead to general schedule delay. ("Things" being design integration, bottlenecks, part supply, resource management, etc. A lot goes into this all!) What would I change for next year to alleviate some of this? First - build a bit sooner, and build a bit faster. Build sooner is hard. We are very CAD-driven, and CAD can help solve a number of design issues before we use time and material in manufacturing. We can absolutely machine faster though. This year, our first iteration of <all> of our parts had lightening pockets. Next year, we can very likely run just hole-patterns, and build the first iteration robot (overweight) more quickly. With physical parts and subsystems in hand, we can test and find more "duh" items faster, expediting the entire shakedown and tuning process. (Also, giving a robot to software while we manufacture more parts.)

Elevator Parts! Week 7!

Second is good documentation. Multiple years now, we have built a prototype, had it perform very well, failed to record the data, and then built a model in CAD that did not follow the measurements of the prototype. (And generally the prototype gets taken apart at some point.) We then go through the full CAD process, manufacture, assemble, and only once we test the final product do we realize that things have changed, and it's no longer behaving the same as the prototype. Nicht gut. (Not good.) Next year - we should follow through on statements we make in pre-season! :-D A prototype should not pass its final review until is has been documented in our internal wiki. Geometry, critical dimensions, power calculations, etc. If you spend the extra time in preparation, you can save double or triple the time in execution.

Formula 1 reference: In 2019, Williams Formula 1 team had to delay their crash test schedule. This delayed their shakedown schedule. Their car arrived at the winter test 2 days late. They were 4 seconds off the pace. (Jeremy Clarkson voice: In Formula 1, 4 seconds is a week!) They developed and improved throughout the year, and ended the last race of 2019 around 3 seconds off the pace, with one World Championship point to their name.

The Ugly

My shoes are pretty ugly. Seriously, ordered them on Amazon to match the CK Red per our branding guide. They're orange. It's... so bad...

Pictured: Not red shoes.

Also our dog-tail side-stepping balls when a belt or pulley comes undone is pretty ugly. Our indexer doesn't like it, software doesn't like it, drivers don't like it, Baby Yoda doesn't like it. Ain't nobody got time for that.

Pictured: Stuck ball. Twisted dog-tail. Not pretty.

Formula 1 reference: Literally any of the 2014 cars... Just... hideous... (I wouldn't Google them on a school network. They're so bad...)


Our strategy team is top notch.

Lead Scout: "That's true, but I think scoring 8 balls will help us more than scoring 5."

Mmm... sounds legit....

-B