pic: Team 221 LLC. - Wild Swerve Module

[R.C.]

Great product guys. I really love the design, there is really no point to debate the issue. As Cory stated earlier, how many people make swerves? I know my team first started out by buying the super shifter and learning from it. Buying off the shelf parts are great for learning. I know my team will buy it and play around with the idea. But I'm pretty sure that my team will stick with 6wd (cough, 8wd was kewl to try). But AJ, that is a super sweet product and best of luck man.

-RC

[Akash Rastogi]

Originally Posted by Jared341

Competition is not a bad thing. For anyone.

This is actually awesome because now we know a consumer based swerve drive is definitely possible. I was just joking before

[Foster]

If I have seen further it is only by standing on the shoulders of giants. -- Issac Newton

Here is a chance to stand on the shoulders of Team 221 LLC that is standing on the shoulders of WildStang. Yes, it's not a "Bolt On and Win (tm)" product, but it's something that a team of bright engineers could do something with. And maybe they will release those details so we can all benefit from it.

If I want to TEACH crab drives, I really don't want to spend hour and hours doing the grunt work of getting the four drives to work. I want to spend the time on the hard part of the coordination and control to make the base move.

R. W. Hamming added -- Mathematicians stand on each other’s shoulders while computer scientists stand on each other’s toes.

[Francis-134]

While this is an interesting development, I think this is not a bad thing for FIRST. In fact, it makes me feel good. The more people with swerve drive means more people are going to be pushed to do something they are not familiar with.

For those of you who think this is bad, I have a challenge for you. Build a better swerve drive. While this looks quite good, we are certainly nowhere near the pinnacle swerve technology. Strive to be better and you can be.

[OZ_341]

Any company I ever worked for made very unique designs, which included countless numbers of "off the shelf" components.
Save your efforts for the creative aspects of your system design and don't reinvent components when there is no need to do so.
It makes good engineering sense and good business sense. Its real world and simply the way many engineering departments operate.

As for the educational aspect, you don't lose anything in buying an AndyMark Shifter as long as you discuss with the kids the details of how it works.

[Lil' Lavery]

Many of my points were said before I got a chance to say them, but I'll expand upon a few of them.

To imply that teams do not learn when using COTS items is simply silly, and wrong. It almost strikes me at those who criticize these parts seem to have no experience using them (or similar products). It isn't simply, "hey, put that there" and have the robot magically work. There is still a great deal of engineering design that goes into the process. This is no more of a complete swerve drive than an AndyMark mecanum wheel and kit shifter is a complete mecanum drive.
COTS items are frequently used in real world engineering solutions, and as Al just said above me, there's no need to reinvent the wheel each and every time. Yes, there are certain lessons you gain when you build something from scratch that you won't get here, but the same applies in reverse.
It is truly rare that any pre-engineered solution will exactly fit your specifics and requirements precisely. You may be able to improve on them, lower their weights, or have to add protection to the components in your design. You may have to design a suspension for this drive for it to be viable in a specific game. You may have to adjust ratios to fit your strategy (and depending on the ratios you need, you may have to engineer and construct a whole new gear housing). You may feel a different motor is the ideal drive motor for your design.
You get into a whole new set of engineering lessons and design thinking when you try to improve and understand a design that someone else created. While many have cited certain designs as "influential" to their own, and some may have looked at very detail pictures, had the designed explained to them extensively, or even dug through the CAD files. But it's a whole new ballgame when you get to actually have a physical copy of the module in front of you to work with and attempt to improve (or find the flaws in it that don't fit your design requirements).
And who says that this has to be used exclusively on competition robots? This could very easily be a great prototyping tool (for either pre-season or early build) to help teams start their software training and experimentation early (as well as start training drivers for the potentially complex task of driving a swerve-driven robot). It can be great for building a sweet demo bot.
And the fundamental argument that this is somehow against the spirit of FIRST because of the allegations that it hurts the learning process are, in themselves, flawed. Learning is an awesome and powerful bi-product of FIRST, but it is NOT the primary goal. Inspiration to become a scientist/engineer and the creation of a culture that values those professionals is the primary function of FIRST. This doesn't hinder that.

[gblake]

Jeez - A lot of posts went up fast here - I don't have time to read every word in each one, so I hope the point I will try to emphasize hasn't been beaten to death already. Here goes.

On a wall in the FIRST Place part of FIRST HQ in Manchester, there is a tee-shirt emblazoned with this quote attributed to Carl Sagan, "If you want to make an apple pie from scratch; first you have to invent the universe." When I spotted it yesterday, I smiled.

So, to anyone who is worried that a COTS swerve module is too much function in one purchase; let me be blunt, "Get over it." Here is why

Until you start mining, refining and smelting your own aluminum; and start creating your own lithography masks for your integrated circuits (that you are going to print on your own silicon wafers); and start manufacturing your own pneumatics and sensors and radios and and paint and .... your objections are likely to fall on deaf ears.

Instead of wasting time objecting that something that was once hard has now become easy(ier); move on to tackling the next harder design/coding/manufacturing problem that you couldn't tackle earlier because your either spent too much time trying to building swerve modules or because that harder problem required swerve modules that you knew you couldn't build.

Take it to the next level folks!

Maybe if you build a sentient and self-replicating, but benign and helpful machine, in the 44 days of build season, you can convince me that you have run out of interesting challenges to use to educate and inspire yourselves and your team mates; but until then, be happy that the world is making progress and what was once hard is now easy!

Sheesh! Sometimes the glass really is half full. This is one of them.

Blake
PS: When/if it gets to be too easy to build a machine that can accomplish a recent/typical FRC game, then I suspect it will be time for someone to unveil a harder game.

[Billfred]

Silly thought: Imagine someone taking these modules, stripping off the swervy bits, and using it as an ultra-compact drive module in some other application. It probably wouldn't replace that 6WD setup most of us have sitting around for most strategies of play, but I bet we've all wanted the extra ground-level real estate at some point or another.

[Akash Rastogi]

Originally Posted by Billfred

Silly thought: Imagine someone taking these modules, stripping off the swervy bits, and using it as an ultra-compact drive module in some other application. It probably wouldn't replace that 6WD setup most of us have sitting around for most strategies of play, but I bet we've all wanted the extra ground-level real estate at some point or another.

We actually did something similar to what I think you're talking about. I shall picture hunt....

This should remain a public link.
http://www.facebook.com/photo.php?pi...0&id=645471640


Each of those 4 drive modules is just bolted on in about a minute each. You could potentially raise and lower the modules depending on how you design it. Am I on the same page here?

[Rick TYler]

Originally Posted by pacoliketaco

I read most of this thread already, and as soon as i saw the picture, i was practically disappointed. I know that is not a very nice thing to say, but i dont see this at all to be in the spirit of FIRST. as far as i can see, swerve drive is very complex, and should only be attempted by teams with the capabilities to make one themselves.

Want to wind your own motors?

Cut your own sprockets and gears?

Build your own computer?

Design your own ASICs? (FPGAs are too "COTS.")

Mine your own copper ore, refine it, and draw your own wire?

There is nothing mystical here. Some parts are in the KOP, others are available commercially, and some have to be custom-made. There is nothing magical about any of those categories, and the appropriate allocation of parts across those categories is whatever the GDC says it is. There is no revealed truth here -- just various mechanical and electrical bits and strategies for their acquisition or construction.

[fuzzy1718]

Too all of you wo keep saying "mine your own metal, make this from scratch yada yada yada" everyone is just trying to say where do we draw the line. If you use the that arguement all you are seeing is black and white, not the gray area in between.
Yes, the GDC prevents robot in a box, but the problem is where do we draw the line on the words "complete mechanism?" You could turn this into a shooter rather easy, but how much of that shooter is already there for you? 1/2... 2/3?

Team 221 this is in no way ment to sound ofensive to you, but when you guys offer a product, simply offer parts not a complete anything. For instance with your guy's chassis why sell the whole rail, just sell the parts seperatly and let teams figure it out. Not only could you guys profit more but it would appease both sides of the arguement. You would be selling the whole thing, but teams could still build it mixing your stuff with others.

Just a thought, hope fully I don't get tared and feathered for it.

[Andrew Schreiber]

Originally Posted by fuzzy1718

Too all of you wo keep saying "mine your own metal, make this from scratch yada yada yada" everyone is just trying to say where do we draw the line. If you use the that arguement all you are seeing is black and white, not the gray area in between.
Yes, the GDC prevents robot in a box, but the problem is where do we draw the line on the words "complete mechanism?" You could turn this into a shooter rather easy, but how much of that shooter is already there for you? 1/2... 2/3?

Team 221 this is in no way ment to sound ofensive to you, but when you guys offer a product, simply offer parts not a complete anything. For instance with your guy's chassis why sell the whole rail, just sell the parts seperatly and let teams figure it out. Not only could you guys profit more but it would appease both sides of the arguement. You would be selling the whole thing, but teams could still build it mixing your stuff with others.

Just a thought, hope fully I don't get tared and feathered for it.

I won't tar and feather you, I'll disagree though.

Just as an example I will use a Joe Johnson's NBD and show what percentage of a mobility system it is compared to the percentage for these mechanisms.

First I have to define what a mobility system is. For the sake of definition I will call it the bare minimum necessary for a robot to move. This means 2 driven wheels and 2 omni wheels, a chassis, associated electronics etc.

TEAM 221 LLC (note that Team 221 is not actually an FRC team, it came out of a FIRST team that Anthony was involved with in the past and is no longer associated with FIRST other than selling parts designed for use in the competition)

2 x CIM motors
2 x swerve modules
2 x omni wheels
Kitbot frame from AndyMark (or IFI depending on preference)
CRio
PD board
Digital breakout board
2 x Victor or Jaguar
Wire
2 x Potentiometer
Programming to control the swerve drive
2 x KOP Wheels


NBD

2 x CIM motors
2 x Dewalt drill transmissions
2 x omni wheels
Kitbot frame from AndyMark (or IFI depending on preference)
CRio
PD board
Digital breakout board
2 x Victor or Jaguar
Wire
Default Programming
2 x KOP Wheels


Seems to me that the Swerve modules are not a significant portion of the drive system. You still have to wire them, still have to assemble them, still have to mount them. Not only that but you have to program them which everyone who does swerve says is the difficult part. Now, in my opinion this is a pretty crappy use of the swerve modules but it DOES show what is needed in a basic mobility system. I could probably assign weights to all of this but they would be highly subjective. Instead I will bring attention to the fact that the Team 221 Swerve actually requires MORE parts than a bare bones set up using Dewalts.

NBD does require you to make some modifications to the Dewalt gearbox but these are all detailed in the white paper so I count this as roughly the same as assembling something based on instructions from an educational experience, I feel this will be the sticking point for many people. Which drive train do people learn more from? To put it bluntly, Team 221 has the distinct advantage here. Programming a swerve drive to work reliably and simply is challenging from a programming point of view. NBD has the benefit of pulling the default code down from FIRST and you are up and running with minor if any changes. Mechanically speaking both teams would learn roughly the same amount assuming neither opened up their parts and toyed with them to figure out how they worked. Electronically the advantage goes to Team 221 again, they get to learn to wire up a potentiometer (or encoder). For these reasons I have to say that the NBD white paper constitutes a higher percentage of a complete mobility system than the Team 221 swerve modules. Furthermore, NBD actually causes students to learn less when assembling it.

Now, my disclaimers.

This is my OPINION, you are welcome to disagree with it and encouraged to debate it but under no circumstances are you allowed to disrespect me or my opinion based solely on your disagreement. I am more than willing to respond to someone who is willing to show me where I went wrong (in their opinion) but will be very angry if you respond by calling me an idiot or any such childish retorts.

[s_forbes]

Nice product, I hope it has the chance to make it onto some FRC robots in the future!

We're one of those teams that doesn't really have enough resources to manufacture our own swerve modules. I'm sure there are other teams like us that would jump on the chance to build a swerve system now that most of the difficult manufacturing is out of the way. There is still way (waaaay) more work to go into a successful swerve drive of course, but it's nice to have one of the tough to build parts readily available.

Our drivetrains have been slowly getting simpler and easier to build over the years, so I doubt we are likely to purchase some of these. It's great that they're finally out there, though.

[Akash Rastogi]

Originally Posted by fuzzy1718

Too all of you wo keep saying "mine your own metal, make this from scratch yada yada yada" everyone is just trying to say where do we draw the line. If you use the that arguement all you are seeing is black and white, not the gray area in between.
Yes, the GDC prevents robot in a box, but the problem is where do we draw the line on the words "complete mechanism?" You could turn this into a shooter rather easy, but how much of that shooter is already there for you? 1/2... 2/3?

Team 221 this is in no way ment to sound ofensive to you, but when you guys offer a product, simply offer parts not a complete anything. For instance with your guy's chassis why sell the whole rail, just sell the parts seperatly and let teams figure it out. Not only could you guys profit more but it would appease both sides of the arguement. You would be selling the whole thing, but teams could still build it mixing your stuff with others.

Just a thought, hope fully I don't get tared and feathered for it.

First of all, Team 221 LLC, AndyMark, or any other manufacturer of such products, do not only appeal to the FRC audience. They are robotics technology manufacturers and should always be remembered as a BUSINESS first.

I don't know how you draw the argument that they can make more money by selling parts, I'd rather pay a little(or a lot) extra if I'm in a time crunch (6 weeks anyone?) and need a swerve module at my door at 3am on a Tuesday (happens to all of us).

I wouldn't tar and feather you Josh, as you've been flamed before on here, but definitely look at things from multiple perspectives. If I were the companies in question, this is the exact direction I would move in; selling full modules. Team 221 and AndyMark are two companies RC and I look up to because, for AndyMark at least since 221 is new, they have a great established business model for the robotics technology world. Even look at VEX, they sell full arms now because that's how you run a business. Just look at all the angles of the situation. Also, if I were them I really wouldn't bother trying to appease both parties of the argument, the one that likes what the company is selling would be my target audience at hand (consumers) and the others would later on potentially see what they've failed to benefit from.

+0.02

I love what's going on with VEX, Team 221, and AndyMark. Keep it up guys. You may not know it, but you really do inspire those of us who may be a little more business minded than mechanical or have the ability to combine both skills, kids like me.

[gblake]

Originally Posted by fuzzy1718

Just a thought, hope fully I don't get tared and feathered for it.

No tar, no feathers. Dang

Hmmm - How about pitchforks and torches???

Originally Posted by fuzzy1718

Too all of you wo keep saying "mine your own metal, make this from scratch yada yada yada" everyone is just trying to say where do we draw the line. If you use the that arguement all you are seeing is black and white, not the gray area in between.

Quite the opposite, all we are seeing is gray. Unless you are going to reinvent the universe on your own, everything in this dicussion is a shade of gray (That is the point of the Carl Sagan statement).

Some might say that a black and white opinion is one that says selling the modules is a bad idea; instead of expressing that selling the modules has both pluses and minuses; and then perhaps expressing a reason or two for why the minuses outweigh the pluses

Blake