Swerve vs. Butterfly Drivetrain

[AlexanderTheOK]

Originally Posted by Jon Stratis

And that right there is what most teams have problems with. Building a "working mock up" that is similar enough to actually let you do what really needs to be done isn't always easy. What works with small, weak motors when supported off the ground doesn't always translate directly to strong, fast motors hauling around 150lbs. Getting the PID controllers set up correctly can be horribly time consuming.

Fair enough, the difficulty of some systems comes from their scale. A servo motor waving a toothpick isn't the best way to test code for the arm of a 2011 robot.

I can say from experience however that the code for a swerve drive does in fact scale rather well. It took me possibly 3 days to port the code over to java, 20 minutes to tune the PID, and another 3 days to make it look pretty so we could debug things, (and likely 4 to deal with the nasty issue of analog counters in the wpilib).

The rest of the issues the system had stemmed from the unbelievable complexity of designing and building the darn thing (of which, to the credit of those who worked on designing and building it that year, there were very few.)

Originally Posted by Jon Stratis

If you can build a working mockup that is sufficiently complex in one day for your programmers to use, why does it take 6 weeks to build the real thing?

I do have to take offense at what seems to be deliberate misreading of my post.

It took 3 days to build the mockup. It did not take 6 weeks to build the whole drive. The base was working by the end of week 4 that season. The first I directly stated, and I believe you should have noticed, the second is rather easy to infer, which I assume you also did.

3 days does seem a bit short though. I would expect a team with fewer vex related components on hand to take more time due to the process of procurement.

Originally Posted by Chris is me

Every team is run differently and is composed of totally different people of different skill sets. Some teams have barely 1 student who can write all of the robot code, some teams have a few students but no mentors, some teams have a mentor or two but no students, and others have entire software teams at their disposal. Exposure to control theory and advanced embedded control is similarly mixed.

A rather good point. I was a bit more experienced at that time than what I imagine the mean would be.

I would argue that a team with the means to build an effective swerve drive is almost guaranteed to have at least one programmer with the experience and talent to figure it out rather quickly,

but such an argument would be based entirely in conjecture since I have experience only with two teams. It would stand to reason that I had not considered teams with a greater imbalance of resources between the programming and mechanical teams.

Originally Posted by Chris is me

Even your own example is of a swerve drive that, in your words, had a configuration that did not work.

Right, seems a component of my post is missing. I was sure I had typed it but voila, as I was going back to look for it it was gone. There was a bug in the wpilibraries. It caused the configuration to not work. What I had failed to mention (but which it would reason isn't too difficult to infer) is that the bug was indeed fixed (good god would I have had a rant to put up here if it wasn't.) in, if memory serves me well, approximately a snappy four days. The configuration itself was fine, and worked, again, if memory serves me well, flawlessly.

[dmorewood]

Originally Posted by itsjustjon

I mean, I'd hate to sound argumentative but all of these drives are far too complex of a solution to the FRC games given in the past five years, give or take.

You'd be better off 86'ing any of these and switching it for a WCD due to easy manufacturing. And, since all of these FRC games are designed with rookie teams in mind, you'd be hard-pressed to find a game where Swerve, Mecanum, Nona-whatevers are necessary for success.

I actually think swerve was a very good solution for 2014 and we enjoyed having it in 2015. Back in 2014 it was a lot harder to play defense on a swerve bot with a good driver then it was to play defense on a WCD bot. This is best demonstrated by 1640, who ended up being Einstien Finalists with 1114. In 2015 we used swerve on our landfill bot and it was exceedingly useful for quickly maneuvering to intake totes and just being able to turn in place to score our tote stack made us less susceptible to losing the stack. Hope these examples help. Personally I will always advocate for swerve on a relatively flat field where heavy defense could potentially be an issue.

[Bryce2471]

As advice for January I have to recommend using a WCD or similar skid steer drive. So if your team has not explored these thoroughly, you should start there.

However, choosing between a butterfly and swerve drive as an off season project is a completely different matter. The swerve drive is by far the better option, for many of the same reasons that it should not be used in most FRC games. Swerve drives are expensive, difficult to design, difficult to machine, difficult to assemble, difficult to program, difficult to maintain, and difficult to drive well. But when you get it all right, Swerve drive is an amazing sight to behold. This means that when you get done building a good swerve platform, your team will have grown and learned in every way, and you'll have a killer demonstration bot.
(Just my two biased cents)

P.S. I am biased because a swerve drive robot sparked huge inspiration in me as a freshman.

[Ginger Power]

This thread has mostly felt like a referendum against swerve drives, which has been telling. Given my current situation and the feedback here, it definitely seems as though butterfly is a better option for our purposes.

Some have mentioned that both drivetrains are unnecessary/too complicated for FRC and that WCD/Kop drivetrain is the better option. The point was made that a simple tank drive will offer 95% of the performance for a fraction of the cost. While that may be true, why isn't it worth investing build season time to implement a more complex drivetrain that has been perfected in the offseason vs. a WCD or kop drivetrain?

[asid61]

Originally Posted by Ginger Power

This thread has mostly felt like a referendum against swerve drives, which has been telling. Given my current situation and the feedback here, it definitely seems as though butterfly is a better option for our purposes.

Some have mentioned that both drivetrains are unnecessary/too complicated for FRC and that WCD/Kop drivetrain is the better option. The point was made that a simple tank drive will offer 95% of the performance for a fraction of the cost. While that may be true, why isn't it worth investing build season time to implement a more complex drivetrain that has been perfected in the offseason vs. a WCD or kop drivetrain?

The biggest problems I think teams face in implementing butterfly or swerve drives in season is a lack of testing pre-season, poor design choices, and a lack of good machine tools. I think octocanum can be done fairly well with a manual mill, but the design has to be very similar to what was done in the offseason to have a driving robot within a few days. IMO well-designed swerve drives are very rare, and often times teams do not have the machines to make them well.
Butterfly + strafe wheel takes up space in the center of the chassis, which is rarely available as well.
If you have the machines, the design experience, and the pre-season testing, I see no reason why you can't do swerve or butterfly/octocanum. Just make sure it actually benefits your strategy and/or driver before doing it!

[Andrew Schreiber]

Originally Posted by Ginger Power

This thread has mostly felt like a referendum against swerve drives, which has been telling. Given my current situation and the feedback here, it definitely seems as though butterfly is a better option for our purposes.

Some have mentioned that both drivetrains are unnecessary/too complicated for FRC and that WCD/Kop drivetrain is the better option. The point was made that a simple tank drive will offer 95% of the performance for a fraction of the cost. While that may be true, why isn't it worth investing build season time to implement a more complex drivetrain that has been perfected in the offseason vs. a WCD or kop drivetrain?

It's a function of priorities for me. If my goal is to be as competitive as possible the goal is to be as simple as possible and get as much practice and iteration in as possible. But sometimes you just wanna build something that's MFD.

[Chris is me]

Originally Posted by Ginger Power

Some have mentioned that both drivetrains are unnecessary/too complicated for FRC and that WCD/Kop drivetrain is the better option. The point was made that a simple tank drive will offer 95% of the performance for a fraction of the cost. While that may be true, why isn't it worth investing build season time to implement a more complex drivetrain that has been perfected in the offseason vs. a WCD or kop drivetrain?

The big thing is, even when you know how to do swerve drive and you've perfected it in the offseason, it still takes more time, resources, and effort than tank drive. The drive still won't work better than tank without great code running consistently, and you are robbing efforts from your manipulator development, drive practice, etc. to do swerve.

Basically, you have to decide that swerve is such a competitive advantage that it is worth putting less polish into other systems of your robot. In most games, it just hasn't been. There hasn't been a game where optimized omnidirectional drivetrains (swerve, mecanum, or otherwise) are strictly better than tank drives, other than 2015. And in 2015, mecanum drive built and programmed well could achieve what swerve drives could without as much mechanical or software complexity.

(note for those following along at home: this is a post where I'm arguing mecanum drivetrains were the best choice for a particular game. Hell hasn't frozen over, has it?)

[Taylor]

Does a butterfly-style drive necessarily imply the use of pneumatics? For a robot design that does not otherwise incorporate pneumatics, it is certainly important to factor in the space, weight, wiring, and programming required for a pneumatic system to support the drivetrain.

-----

Also, it's important to consider levels of implementation. When I think swerve, I think 16, 71, 111, 118, 1640. Those aren't representative of the 'average' swerve drive robot.
If a butterfly drive succeeds, it's pretty nice. If BD fails, it's still a completely capable 4WD robot*. However, there hasn't really been a team that has used it consistently enough, and at a high level of success, to be the standard-bearer for that configuration.

*the same can be said for a failed SD, but mechanical locks may be needed to achieve that.

[Ari423]

Originally Posted by Taylor

Does a butterfly-style drive necessarily imply the use of pneumatics? For a robot design that does not otherwise incorporate pneumatics, it is certainly important to factor in the space, weight, wiring, and programming required for a pneumatic system to support the drivetrain.

-----

Also, it's important to consider levels of implementation. When I think swerve, I think 16, 71, 111, 118, 1640. Those aren't representative of the 'average' swerve drive robot.
If a butterfly drive succeeds, it's pretty nice. If BD fails, it's still a completely capable 4WD robot*. However, there hasn't really been a team that has used it consistently enough, and at a high level of success, to be the standard-bearer for that configuration.

*the same can be said for a failed SD, but mechanical locks may be needed to achieve that.

For butterfly drive you need to physically switch which wheels are touching the ground. This movement requires fairly quick actuation, a very large amount of force at stall (enough to support the whole robot), and only two positions. To me at least, those restrictions scream pneumatics. I guess theoretically you could use motors (big servos maybe?) but that would be way more complicated than with pneumatics where all you need is a cylinder pushing between the module and a hard attachment to the chassis.

IIRC 148 has used variants of butterfly drive in a number of games (I know they used it in 2010). They haven't used it every year like some of the team's above have used swerve, but if I had to pick a team to be the standard for butterfly drive, I would pick them.

[iyportne]

This thread contains many excellent perspectives. And yes, swerve is not easy to employ but worth the journey. We tend to look at swerve as our development infrastructure and learning bed. We are currently on version 4.0 and have successfully used swerve to our satisfaction in the last two seasons. Code is rock solid and and basically a library object so scalability is not an issue. We have a solid working encoder and have now figured out a robust position and mount to survive an entire competition. We use our technology to develop relationships with local machine shops off season so we have that covered (though the last two seasons we used the Team 221 Revolution Pro modules). We have working robots to practice with so driver training is also covered. I would suggest that if you are interested in any advanced drive train(s) that you make it a multi-season effort and manage your competition risk as you go. And also, you don't have to re-invent the wheel (ha-I just saw the pun there) since FIRST rules dictate that if any of us develop off-season hardware or logic, we must provide design and code to the FIRST community prior to kickoff in order for us to use it that season. We post links to our Swerve design models and JAVA code here on CD.

A few more benefits to swerve added to those already mentioned are:
1. You are no longer limited to a two sided functional robot (front-back). Programming wise any side can be designated as front, and can be switched on the fly.
2. You can easily switch between or mix field-centric and robot-centric movements with vector math.
3. Your center of rotation is now virtual rather than mechanical, so it can exist anywhere - even outside the robot perimeter - great for object acquisition or placement.
4. Sure we use 8 motors, but the entire robot is now a very predictable 360° continuous turret while static or in field-centric motion - that means you have a turret for object acquisition and launch/placement.
5. With a full 150 lbs. competition weight on 4 swerve wheels pointed to the center of the robot, you have a pretty awesome brake for resisting defense or for ramps and it remains active at power-off.
6. You have full 4-wheel positive traction 100% of the time (obviously on flat surfaces) during any movement and don't suffer the movement control loss when not all four wheel are in weighted contact with some surface.
7. Our current development has gotten the net drive train weight near or below the standard KOP drive train.
8. If the game requires obstacle negotiations, you don't have to abandon the benefits of swerve, think of other add-on motion devices like belts, cogs or tank treads...remember Swank Drive?

We should have our Strange Swerve 4.0 designs posted before kickoff. Good luck with the development path you choose!

[marshall]

Originally Posted by iyportne

This thread contains many excellent perspectives. And yes, swerve is not easy to employ but worth the journey. We tend to look at swerve as our development infrastructure and learning bed. We are currently on version 4.0 and have successfully used swerve to our satisfaction in the last two seasons. Code is rock solid and and basically a library object so scalability is not an issue. We have a solid working encoder and have now figured out a robust position and mount to survive an entire competition. We use our technology to develop relationships with local machine shops off season so we have that covered (though the last two seasons we used the Team 221 Revolution Pro modules). We have working robots to practice with so driver training is also covered. I would suggest that if you are interested in any advanced drive train(s) that you make it a multi-season effort and manage your competition risk as you go. And also, you don't have to re-invent the wheel (ha-I just saw the pun there) since FIRST rules dictate that if any of us develop off-season hardware or logic, we must provide design and code to the FIRST community prior to kickoff in order for us to use it that season. We post links to our Swerve design models and JAVA code here on CD.

A few more benefits to swerve added to those already mentioned are:
1. You are no longer limited to a two sided functional robot (front-back). Programming wise any side can be designated as front, and can be switched on the fly.
2. You can easily switch between or mix field-centric and robot-centric movements with vector math.
3. Your center of rotation is now virtual rather than mechanical, so it can exist anywhere - even outside the robot perimeter - great for object acquisition or placement.
4. Sure we use 8 motors, but the entire robot is now a very predictable 360° continuous turret while static or in field-centric motion - that means you have a turret for object acquisition and launch/placement.
5. With a full 150 lbs. competition weight on 4 swerve wheels pointed to the center of the robot, you have a pretty awesome brake for resisting defense or for ramps and it remains active at power-off.
6. You have full 4-wheel positive traction 100% of the time (obviously on flat surfaces) during any movement and don't suffer the movement control loss when not all four wheel are in weighted contact with some surface.
7. Our current development has gotten the net drive train weight near or below the standard KOP drive train.
8. If the game requires obstacle negotiations, you don't have to abandon the benefits of swerve, think of other add-on motion devices like belts, cogs or tank treads...remember Swank Drive?

We should have our Strange Swerve 4.0 designs posted before kickoff. Good luck with the development path you choose!

Having watched these guys develop their swerve system as well as borrowed their lessons learned as well as those from Anthony at 221, this is all really good info.

I'm not 100% convinced about mixing other drive types with swerve (swank) but these guys made it work and it was impressive to watch.

Swerve is hard to get right.

[ctt956]

Originally Posted by iyportne

[snip]
7. Our current development has gotten the net drive train weight near or below the standard KOP drive train.
8. If the game requires obstacle negotiations, you don't have to abandon the benefits of swerve, think of other add-on motion devices like belts, cogs or tank treads...remember Swank Drive?

We should have our Strange Swerve 4.0 designs posted before kickoff. Good luck with the development path you choose!

Originally Posted by marshall

Having watched these guys develop their swerve system as well as borrowed their lessons learned as well as those from Anthony at 221, this is all really good info.

I'm not 100% convinced about mixing other drive types with swerve (swank) but these guys made it work and it was impressive to watch.

1533's swank drive from 2016 was definitely impressive. That robot could do everything in the game, and it did it fast. I think the swerve part saved time because of the ability to move sideways/diagonally/faster than a tank drive could make the same movement, if it could do it at all. IMO, it was easily one of the best robots within the NC district for 2016, and still amazing outside of the district. I saw similar designs with other teams(330 is a notable example); not sure whose design inspired whose, or if several teams had the same ideas.

[troy_dietz]

Originally Posted by ctt956

I saw similar designs with other teams(330 is a notable example); not sure whose design inspired whose, or if several teams had the same ideas.

I'm not sure if I'm misunderstanding something, but we had an 8 wheel tank this past season. The last time we did anything other than tank drive was in 2009 with our single module "swerve". (closer to a giant turret with bumpers)

[EricH]

Originally Posted by troy_dietz

I'm not sure if I'm misunderstanding something, but we had an 8 wheel tank this past season. The last time we did anything other than tank drive was in 2009 with our single module "swerve". (closer to a giant turret with bumpers)

And if it weren't for the trailer, it would have been a full turret. Tank drive at heart, though.

The only previous times 330 had a non-traditional tank drive, either they never made the field ('05 mecanum) or they were just an added feature (1999's rotating drivetrain could roll front wheel modules and rear wheel modules independently--handy for dropping down to snag floppies or raising up to slide onto the puck).

330 has never attempted a swerve, other than that '09 robot (which... well, it was a well-camouflaged tank drive). Unless they've tried something in the last couple offseasons, which I doubt.

[XaulZan11]

Especially for 3rd/4th robot, a butterfly drive will likely be viewed as a negative while a swerve will likely be viewed as a positive when it comes to alliance selection.