Preparing for Off Season and Construction season

HII, I hope all of ur teams have had the best season! Now we have finished I want to ask you something

To be prepare to the next season, what are the imprescindible actions and things you made to the offseason and construction season about your engineering area?

About my team, actually our best things we do is to prepare training to our newest and actually participants, prepare a Gantt to our planificación to the season that is every season changing about how it works, made a checklist about our things we have at the workshop and what we need and this season we are planing(still checking) to make an internal offseason, not like the original season but mostly, only where we can do a robot as it was the real season again to the newest participants, and that’s only a little, what do you think and what do your team does?

I’m certain there are people that will disagree with me on this.

Pick your drivetrain style before the game
There is rarely a more compelling reason to choose a certain drivetrain than the following reasons:

  1. It’s the drivetrain the team knows how to build best
  2. It’s the drivetrain the team knows how to program best
  3. It’s the drivetrain the team knows how to drive best

The minor benefits you gain from building a specialized drivetrain rarely leads to substantial increased performance. Even in 2016, many teams were successful with KOP with pneumatic wheels.

Note: From here on out, when I say “custom drivetrain, I mean custom to the game.”

Now, this is the part people start shaking their heads. “No no no,” they say. The custom drivetrain does give use benefits. It gives us an edge over the rest of the competition.

At what cost? How long did it take you to CAD that custom drive when you could have been improving other mechanisms. How much time did your programming and drive team not get to have with a robot?

"The Kit of Parts drivetrain and a custom drivetrain both take 3. The difference is whether it’s days or weeks"

This last year, it took us 3 days to get the KOP driving. After that, our driver and programming team had a robot to work with. We had places to prototype mechanisms. We were able to test things out. It was wonderful.

So how long should drivetrain development be? Short. How long is short? Even shorter. I believe 3 days is still too long to build a drivetrain. If one knows what type of drivetrain they want to build, then all the parts to assemble it should be lying on the table. Code can already be waiting (publicly released of course). Electrical ready. Drive train should be done day one.

Continue Reading

This text is hidden because it starts to diverge from the topic of the thread, but might still be helpful.

Having a drivetrain day one can be critical to the idea that a robot should work every single day of build season. If you test your robot everyday, then you work out many of the bugs early on. There is minimal down time. Most importantly though, it can help you set up a system to optimize for change.

Optimizing for Change
On my alumni team 1747 (2014-2017), we were considered fairly strong. Yet it wasn’t consistent. Some years we were much more powerful than others. I only have realized, 4 years later, what caused this issue.

When I was working with my team, we functioned off a waterfall design philosophy. The game would be analyzed, we build a few proofs of concept, them designed an entire robot, build it, programmed it, and played it, in that order. The robot we came up with was nearly identical to the one we first envisioned. Sometimes, this robot was exactly what we wanted. Sometimes, it wasn’t. We had on years and off years, but were typically above average. The design philosophy was holding us back, but it worked pretty well. Little did I know that lurking behind the corner there a more sinister negative to this design philosophy.

Upon joining 1646, the waterfall principle had ways work against us. Grand designs were made, only to fall through and revert back to the most basics that we weren’t even particularly good at. Way too little time for programming and drive team. We had events where programming was being tested less than 24 hours before.

These things made me realize something had to change. we couldn’t go on functioning like this. That’s also about the time I learned about Agile development.

Agile Development
The basics of agile is the idea that work is broken into short phases (called sprints). Each sprint starts with a planning phase, and ends with a review phase.


These short sprints allow the robot to grow in phases. If any sprint is failed, only a little bit of progress is lost. Then, teams can become stronger over the course of the build season, as well as adapt to possible change that might come along the way.

Starting with a drivetrain day 1 gives you are starting point where all sub-teams can start their sprints. Sprints can be focused on how to improve this specific quality of the robot without losing progress. This way you can always be steadily improving your robot.

How to Pick a Drivetrain
Remember early we mentioned the best reasons to pick a drivetrain?

  1. It’s the drivetrain that the team knows how to build best
  2. It’s the drivetrain the team knows how to program best
  3. It’s the drivetrain the team knows how to drive best

If you want to go with a drivetrain different from the KOP (popular choices being West Coast Drive or Swerve), then in preparation for the build season you should develop one of these drives. Then, before build season, your team needs to honestly evaluate if you which drivetrain accomplishes all the above, and choose to go with that.

"Can you control it as well or better than a 6 wheel tank drive?
If yes, you are ready,
if no, it wouldn’t be as good as a 6 wheel tank drive in season."


Goal Setting
Possible more important that drivetrain is proper goal setting. Mike Corsetto has a great video on Goal Setting. Effective goal setting allows for your team to reflect on how well you did. What went well, and what could we do better next time. Reflections should happen weekly during build season so that you can continually improve.


Know the Basics
If you don’t know the basics of robotics, and know them really well, then you are going to have cracks in your foundation. This is what separates the good teams from the great teams. Yet it’s also the most subtle part about robotics. Rarely do we see it.

Even more concerning is the possibility the you might not even recognize the basics because you know so little about the area.

How does one identify the basics if they don’t even know what they are? I highly suggest using the compass alliance’s pathways to identify if you are missing out on the basics. There pathways are not only great for beginners looking for ways to get better, but also for veterans that want to make sure they have the basics on lock.

5 Likes

If a team uses the same basic drivetrain every year with adjustments for length and width to suit each game, their programmers and drivers could have something to work within 3 hours by using the chassis of a robot from a previous game.

After a few years, the programmers, drivers and team members doing prototyping would all have something approximating the final robot on the first day.

Agreed. I made a comment regarding this:

"when I say “custom drivetrain, I mean custom to the game.”

I’m not sure I properly emphasized that point enough. Any drivetrain, if properly tested ahead of time, could be considered a “standard” drivetrain for your team.

As is likely the case with many teams we are looking at a near total turnover on both build and software since our last Kickoff to Play cycle. At least we got a comp in just before Covid. So there is some but frankly minimal institutional memory regards drive trains.

We have been in serious recruiting mode and have a talented bunch of prospects. We’ll be running three frames shortly. 1. KoP with brushless. 2. Battered but still functional comp robot with H-drive and omnis 3. Swerve drive with Thriftybot units. The latter is down the road a bit yet, one module was assembled in Spring sessions and programming has it in hand. Summer middle school camp is starting to assemble more and has built a simple tube and gusset frame with custom mounting plates.

They will have to decide which is appropriate for the game at hand, but in the meantime we are encouraging lots of drive practice and have given the newbies a simple FRC level action to carry out with the KoP/brushless test bed. Hopefully I can post on this in a week or so…

T. Wolter

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