To Rebuild or Not To Rebuild

“Alas, poor 2020 FRC season! I knew him well” – Hamlet, almost

With Infinite Recharge being replayed in the 2021 season, we inevitably have to deal with the question of whether to build a new robot or just make improvements to our 2020 robot. Personally, I’m still on the fence on what to recommend to my team. We’re starting to talk about which path we want to go down, and I’d like to have a well-reasoned opinion to present. So I started this pros (& cons) list for each option, and I’m hoping to get some help from the community to add to it. Obviously the situation is different for every team and each team needs to make the decision for themselves, but the benefits/drawbacks of each option should remain more or less the same for everyone (though I should note that some of these are pretty specific for my team)

 
Build a New Robot:

  • We get to build a new robot
  • The students get a more “standard” FRC experience
  • We get to learn from our mistakes last year
  • We can make significant changes to the robot if there are rule changes
  • There’s a possibility of using the swerve modules we’re developing in the offseason
  • We can pick a different strategy to play a “new game”
  • We can copy/steal the mechanisms that worked from the 2020 robot to ease the workload

Improve Last Year’s Robot:

  • Our budget will probably be reduced so we can’t spend as much on the robot
  • We already have a competitive robot so why mess with success
  • It gives us a lot of time to improve the small things
  • We don’t know how much time we’ll have to work on the robot during the season
  • Most teams won’t improve their 2020 robot more than we would be able to
  • We can still build new mechanisms to improve robot function
  • We would get to do really advanced control stuff that we usually don’t have time for
9 Likes

Another question that should be brought up is if such a decision should be made prior or after the rule changes get released from FIRST. Depending on how drastic the rule changes are, it has the potentially to drastically shift the opinion of the team.

9 Likes

This is how we’re approaching it too. Prototype, CAD, test things with the existing robot? Sure. But we likely won’t push the button on things until Kickoff (assuming conditions let us push the button at all).

It helps that our robot was semi-disassembled for maintenance and upgrades at the time everything shut down, so we don’t have quite the same sunk cost mindset.

1 Like

As someone from a team that didn’t get to play, I don’t really see a reason for us, or any other team that didn’t play, to do an in-season rebuild unless FIRST allows teams to enter a second robot into 202* competitions.

In the off-season, I imagine we’ll do some kind of project for training purposes, if not a rebuild for CowTown.

Why not both?
It is a fantastic experience for students to build a new robot from scratch, and getting to try things like swerve for the first time will help your team in future years.

In addition you can iterate on the current robot and try out the really advanced control stuff you mentioned.

I’d say that the likelihood of major changes to infinite recharge is low (the point of the replay is to allow teams to use their unmodified 2020 bots), and you can always modify both robots if changes come up at kickoff.

Even if FIRST only lets you enter one robot for 2021 (or if there is no 2021), there will most likely be offseason competitions at some point in the future where you can run both robots.

If my team ever gets access to a build space (unfortunately unlikely for 2021), I’d imagine we’ll try to both iterate on our current robot and build a new one.

2 Likes

ignoring the current state of the world, that’s a fair amount of money to spend that teams may be hard pressed to find.

4 Likes

I think a lot of teams are going to go with the “Play their 2020 robot” strategy, rather than build a whole new robot from scratch. I would honestly be surprised if most teams are able to even meet on a regular basis in 2021, let alone have a “Normal” build season that consists of dozens of students/mentors meeting 5-7 days a week for weeks on end.

Im going into 2021 assuming that I won’t see more than a handful of my team members at a time, and that our first “Full team meeting” (with our 150+ members) wont be until the Fall of 2021. If you’re planning on building a robot from the ground up, you really need to start that process now, and plan for how you are going to design and build given the current constraints. Building a robot in 6 weeks just isn’t going to be possible in 2021.

I’d appreciate it if we could stay away from “my team is doing _______” or “most teams are going to do _________”. Every team is different and is going to weigh the pros and cons differently. I’m just looking to list the pros and cons in this thread to help myself and others make an informed choice. Thanks in advance for staying on topic :slight_smile:

2 Likes

Making improvements in CAD seems a no brainer.

We have been looking at that - harvester and climber in particular.
Our climber was tall after our short one fell apart in testing before week one,
But we think we can make one that works for going under color wheel.
Our harvester would sometimes let the robot drive over balls - bad - we can fix that?

As a team that has used swerve in the past - not this year - maybe we can look at that, as learning opportunity - but all is predicated on what amount of meeting time we can have - a complete unknown so far. good luck to all.
We also are looking at refinements to programming - refinements.

1 Like

Something you may want to consider: have your team go through your week 1 brainstorming again assuming that the game is the same. But this time, you will be armed with the knowledge of how the game is actually played and also the designs that both your team and others came up with. Go through mechanism choices again and see if you would chose the same thing given this new knowledge. But the one consideration you need to add this time around is that you have the opportunity to improve on an existing mechanism. The team may decide that iterating and improving a known mechanism is better that venturing into the unknown with a new mechanism. Also, having a robot that is “done” and available for code optimization and driver practice earlier might be favorable to spending a lot of time back in the build cycel.

The groundwork for this should also include the knowns and unknowns associated with your operating practices. For example, if you know your budget is reduced, that should be included as a constraint that the team needs to operate within. The unknowns would include things like how much time you will actually be able to work on the robot or how many people will be able to be in the shop at the same time, etc.

I think you enumerated most of these ideas in your pros and cons list above. My suggestion is that you let the team answer these questions using the framework of a post-kickoff brainstorming loop.

2 Likes

Building a new robot and iterating on the current robot at the same time:
Pros:
Allows the team to get experience using new technologies like swerve drive, which can help the team improve for future years
Provides a more “normal,” FRC experience (as you said)
Allows more team members firsthand design/building experience (lots of mechanisms to iterate on and build)
Allows complex programming/detailed iteration of the current robot.
Doesn’t endanger competitive performance (you never know how the new robot will turn out, but you have the old robot guaranteed)
Allows the “Best of both worlds,” - you get to practice iterating an existing robot and building a new one from scratch

Cons:
More expensive then other options
Requires regular shop access
It will already be difficult (if not impossible) to get a chance for one robot to compete, finding competitions for two will be even harder.
If there are significant game changes you will have two robots to modify instead of one (this could be a pro if it means there is more opportunities for students to get engaged)

I think the decision should be made based off of how much time you get during build season

4 Likes

We’re aiming to iterate. We like how we did (being in position to potentially dish out a wildcard before the season shut down does that to you) but we don’t like how we got there.

Cue the CAD minions working on how to make things better, at their insistence. We’ll see if the mechanical minions can actually implement a few things later.

We’re effectively shut down from in-person until further notice.

We didn’t get to a competition last year and with school starting in 5-6 weeks we still don’t know if we can meet in person.

Right now I don’t see us having a 2021 season. I see the 2022 season starting with only the seniors having gone through a full FRC season, as freshman. With the combination of a long layoff and a lack of in-person training, there’s a possibility that the students could forget a lot of robotics knowledge.

Simply put, we don’t know enough yet to make a decision. We don’t know how much money we’ll have to spend on a robot. We don’t know what sort of restrictions we might have on our meetings. We don’t know what sort of changes to the game will happen, and how they will impact performance of our old robot.

That said, we do have goals. We want to get the students as much experience as possible, have them designing and building and programming throughout the season. We want them to feel ownership of the robot - this includes rookies who weren’t here last year! And we want them to be involved in the decision process so everyone is onboard with the direction the team takes.

2 Likes

It comes down to the same key points as always, just with different details.

  • What are your resources? Many teams will have less money, time together, or both, but will already have a design that does something close - and access to a lot of other design ideas usually not available when the rules are released.
  • What is the game? The game will likely have the same field elements as 2020, but I wouldn’t be a bit surprised if they get rearranged, have different point balances, or if there were a different game piece.
  • What are your goals? What inspires on your team? Is it winning matches/events, learning something different each year, or something else?
  • What’s your strategy? Both for your goals and for playing the game, if different.
2 Likes

It is hard to decipher from the rumors and facts but pretty sure first confirmed the field and game pieces won’t change. It sounds like the game will change very very little; more like a cheezy champ change than moving the center structure. Also the weight limit may increase by like 5lbs to allow for mods.

2 Likes

To decide what we are going to modify in our robot we first went through an analysis using Ishikawa diagram with that we are able to identify the root cases of our problems.
And for us the conclusion was that we wouldn’t need to build a new robot to solve its performance problems :wink:

1 Like

It also feels like “let’s do both” as the answer to “new robot or iterate” feels like a meta-version of the “let’s build a robot that specializes in all the tasks” approach. That’s one that a few teams have the resources and skill to manage, but not many can pull off.

1 Like

I’m going to get this out of the way: I feel like FIRST is going to delay the season or something like that given the current situation.

Anyways, I feel that now is a better time than ever to improve CAD skills for the upcoming season regardless of whether or not a team is rebuilding their robot. I see here that many people are using this summer to prototype certain aspects of the bot.
Our team is having a virtual mock build season of sorts right now to practice strategy and design in groups, which I feel improves team dynamic by making sure that everyone’s voices are heard, not just the captains and increases reliance on CAD members, motivating the team to make communication with them more efficient. Our decision to rebuild or improve will be based off of what happens because of this mock build season.

Also, if a “normal” season comes, it’s inevitable that some of the more organized teams will do some major overhaul to the previous bot or remake the bot: the 2021 bots will be significantly better anyway and an experienced team will know better than being complacent and bringing an unchanged 2020 bot.
It’s like if in a normal season a team performs really well and, say, is one of the only teams with a climb week 1 and attains a very high rank in a district, but then becomes overconfident, and so at district champs, they see that other robots have a better climb while their climb breaks, and thus they don’t go to Worlds. This same principle applies between 2020 and 2021.
Thus, my suggestion is to well-resourced teams is to either rebuild completely from the ground up or at least do some MASSIVE improvements to the existing bot (i.e. disassembling at least 75% of the bot and smashing the pieces together in a manner that gives a much better outcome. My definition of a “normal” amount of rebuilding consists of rebuilding about 45% of the bot).

As one of our mentors have said though:

Then again, I feel like the season will be delayed until a point at which making a bot in 6 weeks will be possible.

Edit: fixed little words at certain intervals to make big sentences clearer in a process called revising

2 Likes