Mecanum vs. Swerve Drive

[mega900997]

I'm doing some research for an offseason project this year. We have used mecanum wheels in the past to great success, however I want to know if a swerve drive would serve us better.

What are the main differences between mecanum and swerve drive and the pros/cons of each drive systems? Or if there is already a thread that discussed this, can someone link it in a post?

[Ether]

There are dozens of threads and hundreds of posts on this very subject. You can use the built-in forum search feature.

Or Google works too.

[jbsmithtx]

Prepare for the worst.

Nearly everyone on CD will tell you that mecanum wheels are completely awful, and have no pushing power.

That being said, there are several benefits to moving to swerve. It is much more efficient in power delivery, and so much more powerful. As a beginning team, mecanum is great for preliminaries. However, as a team that has used them for several years, it may be useful for you to transition into something else. Keep in mind, that swerve is always a work in progress. The best swerve drives are made over several seasons. But don't be afraid to continue with mecanum, if you determine your time should go elsewhere.

Pros to Swerve:
Several alternate control methods
Better power delivery (Speed and no slickness of rollers)

Cons to Swerve:
Not easy to slap onto a robot like mecanum
Requires the use of many motors
Can be unreliable, unless you spend a lot of time getting everything right
Can be expensive (but mecanum can be too)

Pros to Mecanum
Easy to implement
Reliable
Little development

Cons to Mecanum
Rollers
Costly
Can be unbalanced

[Ether]

Originally Posted by jbsmithtx

Pros to Swerve:
Several alternate control methods

What "alternate control methods" for swerve do you have in mind that cannot also be used for mecanum?

Originally Posted by jbsmithtx

Pros to Swerve:
no slickness of rollers

Mecanum rollers do not have to be slick.

[jbsmithtx]

Originally Posted by Ether

What "alternate control methods" for swerve do you have in mind that cannot also be used for mecanum?


Mecanum rollers do not have to be slick.

I meant control methods such as "car" steering, crab, swerve, etc.

And I meant the rollers were "slick" simply because they are a roller...

[buildmaster5000]

Link to a thread about this very subject. While it's been a few years, feel free to PM me with any questions you have. My team did implement a swerve back in 2011 (My senior year of HS) and I like to think I remember a thing or two from the good ol'e days.

Originally Posted by jbsmithtx

I meant control methods such as "car" steering, crab, swerve, etc.

And I meant the rollers were "slick" simply because they are a roller...

Theoretically, any movement that you can make with a swerve you can make with a mecanum, and vice versa.

[Walter Deitzler]

Why not build both?! Swervecanum drive!

Serious: We have built Mecanum. We loved the maneuverability we got, and ease of use. Yes, we could get pushed a lot, but the trick to that was to just go sideways around the defender. They are awesome to show off at science night and such, because everyone like a robot that moves like a crab, and you get questions about the wheels, how they work, etc, from people just looking at the bot.

We do have reason to believe that they we one of the reasons we failed to make eliminations at all in 2012, mostly because there is a stigma against them, and, if you are not first round pick, then there is a good chance you will be skipped second round.

Have fun choosing!

[Pault]

I'm not an expert on this (or just about anything, really), and I am going to echo Ether in saying that there are a million other threads about this already. But I just want to say 2 things.

1. People, please don't let this devolve into another thread arguing about whether mecanums are good or bad. Just stay focused on how mecanums compare to swerve. I think everyone can agree that swerve > mecanum, so lets keep this a simple discussion on why that is.

2. To the OP: don't just dive into swerve like nothing thinking that you will have instant success. Know that swerve is very complex and resource intensive, so be cautious. It's good that you want to test out these kinds of things in the offseason, but know your limits. Basically what I am trying to say is: don't underestimate the difficulty of swerve.

[Samwaldo]

We won the Boston Regional in 2011 with Mecanums, and we were playing the DEFENSIVE ROLE!!! If you know how to use them well, and have a full 120lb robot, defense is very possible with mecanums.

[Boe]

while as a drive system alone swerve always wins it is a lot more difficult and resource intensive and hard to program. Ive hated and thought meccanum drives were a bad idea for a while, but i believe that's because ive never seen a team use them to their full potential. Last weekend i got my first real chance to see KnightKrawler 2052 driving (we go to opposite MN regionals typically) and i was very impressed with their driving and how they used their meccanums to their full potential. I believe KnightKrawler has used meccanums for years (i believe one of my friends said almost every year except 09) and have really figured out how to get the most use out of them. Like all drivesystems the key to using it successfully is driver practice. You cant drive a meccanum like a tank drive which i unfortunately see a lot of teams do.

[Ether]

I think this may be some kind of record. We got through 9 entire posts before someone mangled the spelling of mecanum.

[Siri]

I'll stick this here since the video wasn't out for the other threads.

Originally Posted by Ether

Mecanum rollers do not have to be slick.

True, but I do sort of revel in being able to 'pick up' mecanum bots and drop them on the other side of the field. You know we love you, Jesters.

We're happy to use swerve to push, but more often we use it to run fast cycles, which can't be intercepted the way mecanums are, and toss in a little jostling of nearby opponents. Swerve basically means the above plus that and that (sorry, STORM) as well as this this, this, and this.

Things like octocanum* can theoretically (Ether?) move similarly, though most people seem to concentrate on strafing. But they can't push or prevent being pushed while doing it, which we do rather constantly. I've never met a mecanum--or octocanum, if I put them in the right situation--that we can't push. Most basic tank drives will go where we want them to, and we can at least slow down and vector most of the crazy ones off course. It's better at two-speeds, you could check with 1717.


In short, swerve gets you from A to B while drastically minimizing the number of people that can...redirect you. That said, it's no cakewalk. It's the best investment we've ever made (we've been making it since summer 2009), but it is an investment.

*Switching traction & mecanum wheels. Lots of threads if you don't know the details.

[pmangels17]

Have you given any thought to a crab drive? We did one in 2011. It is similar to a swerve except that instead of spinning each wheel individually by a steering module, pairs of two or in some cases all four wheels are spun by one gearbox. The two pairs method makes turning easier.

You could also simplify a swerve in an easy way. Drive all four wheels from one center gearbox with chain, then spin each wheel with individual steering in order to control direction. This method allows the use of two, three, or four motors, or as many as you choose, and then you can more easily impliment a two-speed drive with your swerve.

Side note: In our 2011 crab base, we steered the front wheels in tandem and the back wheels in tandem. Then, we drove the left wheels with one gearbox and the right wheels with another. We used custom shifter boxes to our advantage. So. much. power. I highly recommend using shifters with common gearboxes.

You could also look at one of the more notable swerves this year, Team 1640. They have four individual modules, one for each corner. I believe they use a CIM on a single speed gearbox and a separate steering motor for each corner. I could be wrong, but that is how they explained it to me at Hatboro-Horsham. Seriously, look at them, they made it to Einsteins, and their drive is awesome. And remember, it takes a different mindset in the driver to handle a swerve versus a crab, though from mecanum they are pretty similar.

[efoote868]

Originally Posted by mega900997

I want to know if a swerve drive would serve us better.

I'd recommend that you develop your own swerve drive in the off season, mess around with it and interact with other robots to get a feel for what you're dealing with.

In my humble opinion, swerve tends to have many more opportunities for failure than other traditional drives. This can be a good thing if you view failures as learning opportunities, or a bad thing if you want to compete with your robot on the field for every match.

Some of the most successful and inspirational teams use swerve drive successfully every year, but their years of experiential learning cannot be replicated in just one season.

Whatever you do, just don't say we didn't warn you .

[mega900997]

Originally Posted by Ether

There are dozens of threads and hundreds of posts on this very subject. You can use the built-in forum search feature.

Or Google works too.

Thanks for this.

Originally Posted by pmangels17

Have you given any thought to a crab drive?

I honestly hadn't thought of that, thanks.

Originally Posted by efoote868

I'd recommend that you develop your own swerve drive in the off season, mess around with it and interact with other robots to get a feel for what you're dealing with.

Whatever you do, just don't say we didn't warn you .

Haha, consider us warned, we don't have any plans on using swerve drive for competitions in the near future but more to get new members interested and for a challenge for other veteran members and it can't hurt to create a showbot using swerve.

We are planning on having our members learn about and how to drive each drive system and for also the drive team to practice for competitions.