Kickoff Practice

My team has always been one of those rural teams that always just made up the robot as we went along. While this led to some creative designs, those weren’t always the best (or even functional). In light of this, our lead mentor, team captains, and I are trying to schedule out build season better, including a goal list and dedicating a design and prototyping stage before we even buy the parts.

My lead mentor took me aside recently and told me that, in December, my team will be using a couple practice sessions to simulate the first couple of build season. They are going to be watching the broadcast from a past game (before 2017), making a goal list, and prototyping a robot design. They will then watch a competition with a robot with a similar design to see how well the idea would work (no word on how they’d simulate driver practice).

I wrote the previous paragraph in third-person because I was assigned to pick the game, watch their process (I can’t interfere at any point), and pick that similar robot. That’s why I came here; I wanted to get some feedback regarding:

•What game should I choose?
•How would I simulate/decide their amount of driver practice? (Low resource, can’t spare a robot)
•Other stuff I can’t think of

Also, none of them browse Chief Delphi (one occasionally asks a programming question), so don’t worry about them seeing this.

1 Like

2009 Lunacy. Because your team should have experience in frustration.

8 Likes

I’m a fan of 2008 for kickoff practice. It had a number of ways to score, each requiring different mechanisms (or lack thereof) and priorities.

2010 is a good year because drive train choice and order of priorities mattered
2012 is fun because it was fun
2013 was a weird game object and it had a few major objectives to go for
2015 was a great design challenge, but not really a game

As for simulating - there used to be a “video game” of each game, I can’t remember what it was called or where you could find it, but that is likely the best way to simulate driving. Though honestly, the prioritizing and prototyping phases are the more important things to practice.

4 Likes

Are you talking about Autodesk Synthesis? That would still require having a robot CAD, but it wouldnt be a bad way to see how a similar bot with a compatable CAD would do.

I also agree with Katie’s years. Find a good game that has a more unique design (any even year game since '08 would be good since none require elevators which have been common the past 2 years and 2015, but any game from 2012 on will yield way better match footage) to really force some creativity and strategic decisionmaking. 2016 is a “good” recent game for strategic decisions on what to pursue and choosing which objectives to ignore, but as i said before any of Katie’s are good choices.

With the time scales it sounds like your team isn’t actually going to be building anything more than rough prototypes, correct? If so then I don’t think you have to worry about driver practice, since there won’t be anything to drive. Just look at how each version of the proposed robots (or robot mechanisms) functioned via videos to evaluate.

As for the game I’d go for 2016 Stronghold because there’s a lot of variation to that game, or if that’s too recent then 2010 Breakaway.

As for other stuff:

  • I’d try to make sure that things stay relatively broad strokes as to what the ideas are.

  • Have them look at the point distribution and assign priorities to which parts of the game are the most important, then you can look at a breakdown of points scored and see if they were right.

  • If someone cheats and goes online to check what was successful in that game only give them a slap on the wrist, because when the Kickoff comes that’s exactly what you want them to be doing: Looking at past games and identifying what mechanisms worked well and how they can adapt them to this year’s game.

We’ve done mock kickoffs for: 2005, 2006, 2007, 2009, 2011, 2013, and 2014.

I’d recommend 2013 and 2016 as good games that require strong strategic analysis.

3 Likes

I highly recommend you use Spectrum’s resources here for mock kickoffs:

@AllenGregoryIV can you update that page for recent mock kickoffs?

3 Likes

Also, if you’re having trouble with “functional” you may want to take the time now to adjust your sails on how you build. Spectrum’s MCC guide helped us make massive progress this year towards having competitive robots, and even if you end up with something more than an MCC it’s still chock full of useful timesavers.

3 Likes

I would recommend picking a goal for where you want your team to place at your events before watching the practice game reveal. After the game reveal you can decide how many features your robot will need to meet that goal. It’s almost always fewer features than you think you need…

Your list of required features will widely vary if your team’s goal is to win a week 1 regional vs making it to Einstein field.

Try not to fall into the trap of thinking “We just needed a 4 cube scale auto and a buddy climb, just like I predicted from the reveal video. Look at how great 254 did with that, great analysis everyone”. Your team should try to get into the mindset of “What’s the absolute minimum amount of robot we need to make our team’s goal? We just want to make play-offs so we probably don’t need a buddy climb”. THEN you can go and watch videos and check if your theoretical bot, with reduced features, was good enough to meet your goals.

Yup, great advice here.

In addition to a mock kickoff, your team should really do a primer using Karthik’s Effective FIRST Strategies presentation.

That must be watched and discussed before any type of mock kickoff happens.

3 Likes

If you want to simulate driver practice, you could always use last year’s robot as is and basically practice cycling using the layout of that year’s field, with the time it takes to do those things in mind (like shooting a ball with a flywheel would take about a second for the shooter to spin up and then you could “fire”)

A great complement to Karthik’s video(s) are the “Strategic Design” videos from Mike Corsetto of 1678.

1 Like

We have been doing our mock kickoffs differently every year and they don’t make as much since to publish. This season we aren’t doing it at all and instead will host a different workshop on FRC robot design, incorporating some of the things from the MCC guide that @Billfred mentioned and suggesting that teams build based on available robot concepts like 118’s Everybot or similar and improving that during the season. Starting with a generally working concept and iterating on it to improve it or add features is a much better strategy for young teams and does a better job of actually representing the engineering design process for most students.

2 Likes

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.