How do Twin Robots Work?

FIRST General Forum
1

I’ve always been curious about this but never thought to ask about it directly. There don’t seem to be any existing threads or discussions about it.

My question is regarding separate teams that build identical robots (in collaboration). 254 and 1538 did it in 2009, 179 and 1251 in 2014, 48 and 3193 did it for a number of years up through 2014, and there are countless other examples.

I would like to know about the logistics of a process like this. Some of my questions:
How do two (or three, 1114/1503/1680 from 2006) teams approach each other about a project like this?
How do you account for the difference in leadership styles, team structure, season schedule, etc. between the two teams?
Do you have to be located in the same city or very close to each other? Do you build in the same location?
If not, how do you make sure that members of each team are equally represented in the design/decision process?
What collaboration tools do you use? Especially for CAD.
How do you ensure that you are designing within the manpower and financial limitations of BOTH teams?
How is prototyping done? Is that divided up (i.e one team prototypes a shooter, the other an intake) or done redundantly?
How is manufacturing handled? Do sponsors make parts for both teams? Do the teams divide the workload (e.g. one team does the milled parts the other team does the turned parts)?
Is the programming up to each team or is that collaborated on as well?
Does the collaboration end once the competition season starts (i.e. do the robots diverge from each other as iteration is done, this seemed to be the case with 48 and 3193 for instance)?

I understand that the answers to most of these questions will differ on a case-by-base basis, but I’m just looking for first or second-hand stories about these projects. If possible, reference the teams involved and post pictures of the finished project if you have them!

As a bonus, I would also be interested in hearing about teams that register more than one team number and build two (identical or otherwise) robots (e.g. 1241/1285 and I think 494 did/does that as well).

11 Likes
2

Onshape

5 Likes
3

Onshape.

5 Likes
4

“JVN’s Build Journal 2010” from 148’s Resource Page has some interesting insights on the interactions between 148 and 217 and some of the design decisions and season progression that collaboration led to.

5

GrabCAD

2 Likes
6

I don’t think Onshape was around during the times of most of these twin robots. I don’t know much about their histories, but I dug up some old threads about the 1114/1503/1680 Niagara triplets (who I’ve always found fascinating) that shed some light on the reasoning/process:

And particularly this post from Karthik: [moderated] pic: The 2006 NiagaraFIRST Triplets!

2 Likes
7

They may not of back then, but something tells me their teams did very recently using Onshape.

3 Likes
8

Pinging @nuggetsyl, @Howard_C, and @Kristian_Calhoun to comment on 25 + 103 collaboration back in the day. Though I was only on the team during one of our twin years (2011) they build pseudo twins in 2012 and nearly identical twins in the 2013 season as well. I know very little about the entire process.

9

All you young’uns are missing the original collaboration. I blame this one for inspiring NiagaraFIRST.

[/geezer]

There was some more generalized discussion of collaboration in another thread that got heated. (note: The rules are MUCH different now.)

2 Likes
10

I think you mean 968. I was involved on the 968 side during 2007-2010, all of which were collaboration years with 254. I’ll say that it really takes the right group of people. It takes people that have similar knowledge, similar design philosophies and ideologies, and personally connections outside of just FIRST Robotics. Even then, it was rigorous and tough for both sides involved. It was a heck of a lot of work and never easy, but there were definitely advantages (and disadvantages) for each team in terms of manufacturing resources and whatnot.

From a technical standpoint, back in the day we did a lot of WebEx meetings. Lots of phone calls. Sometimes a video conference with webcams. I think at one point we had an SVN server for CAD. Back then, 968 and 254 both would do prototyping and CAD. 254 would do most of the precision machining. 968 would do all the jigging, welding, and powder coating, axles and gears. Each team would make 4x quantity of half of the robot essentially. When we both had parts done, we’d call each other up. 254 would drive south, 968 would drive north, we’d meet up in San Luis Obispo, swap parts, turn around, head back, and keep building. Many late nights. It was nuts, and I probably wouldn’t recommend it for teams at a distance, even with modern tools like Onshape and Slack. It’s a lot to manage.

11 Likes
11

There are many 254/968 posts/threads on CD if search for them.

12

He was ON 968 during the collaboration. No thread required.

3 Likes
13

Thats an easy one for us in 2020. 359A and 359B and they are going to different competitions. :slight_smile:

1 Like
14

My comment was more so for the owner of the thread.

15

Yeah, but didn’t you have some sort of collaboration with Wave one year?

16

Yes, 2017 Robot were twins somewhat. And actually played together off-season in China.

17

@juju_beans

unknown-16
unknown-16861×700 268 KB

unknown
unknown769×634 449 KB

8 Likes
18

OK, now for a serious answer.

3863 Pantherbotics and 3647 Millennium Falcons did “twin robots” this year, which was overall fairly seamless thanks to Onshape.

The way we approached the project was largely through our mentorship group. With the overall decrease in CAD proficiency of both teams, @juju_beans and I contacted each other to discuss the possibility of working in the same CAD document this year. On both our parts, we thought it was feasible enough, so we pitched it to our teams. Our teams agreed to collaborate with one another, understanding that we would have to hold each other accountable for CAD deadlines, etc. As we are both small teams, leadership style did not hold much weight in the discussion of twinning. All subsystem leads and technical directors understood that they had equal weight in our discussion - that no one team had more of a voice than the other.

Considering 3647 is in Torrey Pines, and 3863 is in Thousand Oaks – around 200 miles of distance between us two. Neither one of us wanted to manufacture four sets of parts and then meet halfway in LA traffic to swap parts. We simply decided to allow each team to determine the suitable manufacturing processes for themselves, using their local sponsors or machine shops to fabricate their own parts. No physical parts were shared between the two teams.

Prototyping between both teams was a critical element of our design process this season. We largely divided the prototyping between teams, but prototyped in parallel with critical subsystems like the shooter to cast a wider net to see what worked. Contributions from both teams made it onto the final CAD. Pantherbotics was largely responsible for the feeder mechanism and some parts of the intake, while Millennium Falcons helped a lot with the buddy climb and shooter.

This season was not easy for either team. With a desire to be a top-10 team in California, our goals and expectations were set high from the beginning. Even though some functionality was eschewed (going under the trench), we still found everything to be a packaging, manufacturing, and logistical nightmare. 3863’s manufacturing fell behind earlier in the season than 3647, leading to us not having a completed robot until very recently, not even thinking about practice time. Ultimately we both set goals of completing the robot “before the first competition” though, so it seems like it worked out for both of us.

Looking ahead to the rest of the season, we plan on continuing our CAD collaboration in the development of a new shooter. Although we are time constrained between LAN and SDR, we know we can implement a new shooter subsystem while improving our autonomous modes.

Please let me know if some questions went unanswered. I or @juju_beans would be happy to respond!

-Aaron

7 Likes
19

Thank you! Great answer to my questions!

20

1241 and 1285 are a senior and junior team, respectively, at the same high school. Typically, grade 9s and 10s are on 1285 and 11s and 12s on 1241. We share the same build space and mentor pool, although most mentors are linked to one team or the other.

This structure is meant to double the opportunities for students to participate, which I think is really valuable.

Since 2015, our robot designs are usually discussed in separate rooms during kick-off hence why we don’t have identical robots (as well as the fact that we have less experienced students on one team and more experienced students on the other).

1 Like