mecanum vs. omni

[tseres]

o ok....now i get it....so when the robot isnt facing you it still steers like it is.....ya...that would be cool

[Lil' Lavery]

Originally Posted by tseres

o ok....now i get it....so when the robot isnt facing you it still steers like it is.....ya...that would be cool

In addition, field-centric steering is virtually the only way to achieve both rotational and straight lateral motion at the same time ("frisbeeing" in a direction). If the controls are robot centric, the rotational motion will result in the translation to become an arcing motion, rather than a straight line.
See one of the programming related threads I posted earlier.

[tseres]

on that note, it would probably be difficult to program....

[The Lucas]

Originally Posted by 4throck

The one real problem with these drives that we discovered was victor bias. The values that come out are not symmetric, linear, or normal by any means. We ended up using look up tables, ask Kelly for more information about how we did it.

The "center" of the Victor 884 range is NOT 127. It is more like 132 or 133. Some example ranges are provided in this thread in the IFI Forum. I suggest testing your own speed controllers by giving them steady known values (from a pot or wheel and reading it on the dashboard) and looking at the lights.

This makes reversing the motor direction in software difficult, so I suggest just simply reversing the polarity at the Victor output. The motors will still have different power output in forward vs. reverse (that is inherent to the motor) but at least the input electric power will not be the problem. Wheel encoder and yaw rate sensor feedback can correct the former problem.

Here is another good programming link for Jester Drive

[Alan Anderson]

Originally Posted by Alex.Norton

...I would suggest holomonic much more than mechanum.

Drive systems using Mecanum (note spelling) wheels are holonomic (note spelling). The only real difference between Mecanum and omniwheel systems is in the orientation of the axles. Nearly everything else can be analyzed identically. For example, I don't see how a Mecanum wheel sees greater side forces than a traditional omniwheel. Can you explain your reasoning more completely?

[EricH]

Originally Posted by Alan Anderson

Drive systems using Mecanum (note spelling) wheels are holonomic (note spelling). The only real difference between Mecanum and omniwheel systems is in the orientation of the axles. Nearly everything else can be analyzed identically. For example, I don't see how a Mecanum wheel sees greater side forces than a traditional omniwheel. Can you explain your reasoning more completely?

I think I see where he's coming from.

Mecanum:

/// \\\ or \\\ ///
\\\ /// /// \\\

depending on how easily you want to turn.

Omni (45-degree setup):

----
l l
l l
l l
----

The mecanums are always pushing, but a lot of the force is directed sideways by the rollers. However, omni's are often off, so their rollers take the load. The pushing omni's are only pushing one direction. For a mecanum, they are pushing in two directions, unless the robot is moving diagonally.

Some other things to take into consideration: how easy is it to push you? Omni's are a little easier (two-wheel drive in any given direction save diagonal). Mecanums may be set up in two orientations ("easy-spin", which looks like an X, and "hard-spin", which looks like an O) and are harder to push because you have to fight all four wheels.

Ease of building: Mecanums are easier to build. Given four AM planetaries, a set of mecanums, and a kit of parts, it is possible to put one together other than programming in about an day. Omni's need a custom(ized) frame.

Maneuverability: there is a slight difference (I think). Mecanums may be programmed to allow turning while strafing, enabling large arcs. I'm not sure an omni-wheeled robot can do that without spinning.

Traction: Most teams that use mecanums have relatively high-traction rollers (often rubber). Most teams that use omnis do not (often plastic). It's a little harder to push a mecanum robot partly due to this and partly due to four-wheeled resistance.

[JamesBrown]

Originally Posted by Lil' Lavery

It depends on the style on control you want really. For a robot-centric drive (push up on the joysticks, the robot moves to its "front"), a gyro isn't always necessary, but for a field-centric (push up on the joystick, the robot moves away from the driver), a gyro, accelerometer, and/or other sensor packages are almost required.

The gyro can also be convenient in robot centric drive. With gyro feedback you can help to ensure that the robot is actually driving straight when you want it to or that it is rotating when you want it to. The same thing can be accomplished with encoders (or gear tooth sensors on the wheels). As a general rule closed loop control is better than open loop.

[Lil' Lavery]

Originally Posted by EricH

Some other things to take into consideration: how easy is it to push you? Omni's are a little easier (two-wheel drive in any given direction save diagonal). Mecanums may be set up in two orientations ("easy-spin", which looks like an X, and "hard-spin", which looks like an O) and are harder to push because you have to fight all four wheels.

Ease of building: Mecanums are easier to build. Given four AM planetaries, a set of mecanums, and a kit of parts, it is possible to put one together other than programming in about an day. Omni's need a custom(ized) frame.

Maneuverability: there is a slight difference (I think). Mecanums may be programmed to allow turning while strafing, enabling large arcs. I'm not sure an omni-wheeled robot can do that without spinning.

Traction: Most teams that use mecanums have relatively high-traction rollers (often rubber). Most teams that use omnis do not (often plastic). It's a little harder to push a mecanum robot partly due to this and partly due to four-wheeled resistance.

A mecanum is really a form of holonomic drive (as mentioned earlier), but instead of angeled wheels with straight rollers, a mecanum has angeled rollers on straight wheels. Instead of applying a large portion of the force from not a large number of motors, mecanums apply a smaller portion of force from more motors.

As for your notes:
Pushing:
Assuming both sets of wheels have identical traction, it is really the same. You can push on each from angels where you will get partial resistance from all the wheels, or locations where you will ignore 2 wheels (as you will be pushing along the rollers) and fight the full resistance of 2 wheels.

Ease of build:
Mecanums will adapt easier to a "traditional" (kit of parts) frame than a holonomic system. AM Planetaries would be a poor choice for a drive train though (given current rules). The AM CIM motor planetary is more for manipulator use, with a 181:1 reduction.

Maneuverability:
Both Mecanum and Holonomic drives can drive in arcs, spin in place, translate in any direction, and "frisbee" (rotate while translating in a straight line) given proper programming.

Traction:
In general, neither system gives a great amount of traction (because of the general principle of rollers), and each wheel has a direction it can be pushed by (in theory) only overcoming the rolling resistance of the rollers. AndyMark sells both plastic roller omni-wheels, and the "trick wheels" (which you can have rubber and neoprene rollers, with higher traction).

[EricH]

Originally Posted by Lil' Lavery

A mecanum is really a form of holonomic drive (as mentioned earlier), but instead of angeled wheels with straight rollers, a mecanum has angeled rollers on straight wheels. Instead of applying a large portion of the force from not a large number of motors, mecanums apply a smaller portion of force from more motors.

I understand what a mecanum is; my team worked with them between 2003 and 2005 but decided they weren't worth it. My "sketch" indicates only roller direction, and not wheel direction (the axis is along the line)

Originally Posted by Lil' Lavery

Pushing:
Assuming both sets of wheels have identical traction, it is really the same. You can push on each from angels where you will get partial resistance from all the wheels, or locations where you will ignore 2 wheels (as you will be pushing along the rollers) and fight the full resistance of 2 wheels.

With a traditional omni, those locations are much easier to hit. Also, you now have a 4v2 motor battle. 2 can't beat 4 very easily in pushing power, although it can be done. (Assuming that both robots have 4-motor drivetrains). With a mecanum, if you hit it just right (only four locations that are very small), you get less resistance, but it's easy for said mecanum to spin away.

Originally Posted by Lil' Lavery

Ease of build:
Mecanums will adapt easier to a "traditional" (kit of parts) frame than a holonomic system. AM Planetaries would be a poor choice for a drive train though (given current rules). The AM CIM motor planetary is more for manipulator use, with a 181:1 reduction.

I just threw the AM planetary in because I didn't want the BB. Any gearbox will do.

Originally Posted by Lil' Lavery

Maneuverability:
Both Mecanum and Holonomic drives can drive in arcs, spin in place, translate in any direction, and "frisbee" (rotate while translating in a straight line) given proper programming.

Then the question is, which is easier?

Originally Posted by Lil' Lavery

Traction:
In general, neither system gives a great amount of traction (because of the general principle of rollers), and each wheel has a direction it can be pushed by (in theory) only overcoming the rolling resistance of the rollers. AndyMark sells both plastic roller omni-wheels, and the "trick wheels" (which you can have rubber and neoprene rollers, with higher traction).

Traction is not very good, but most omnis that people use seem to not be AM, and of the AM's I've seen, very few are "trick". Most mecanums are a) custom built or b) AM, and both (tend to) use sticky material. The thing is, what role do vectors play? for each wheel, you have two vectors. One is in line with the wheel; one is at an angle. Those vectors govern how much traction/force is applied in any given direction. Mecanums can apply a lot of force going sideways, but a standard omni can too.

[joking, almost]I have an idea: let's settle this. 357, 1138, and 1540 (mecanum) in a series of tests against 116 (2005) and two other onmiwheeled robots (maybe 67 2005 for a three-wheeled version?). Tests include maneuverability, simplicity, and traction (old-fashioned shoving match, choose your direction of push). The better system is the one that wins the most tests.[/joking, almost]

[efoote868]

Word of advice: Build the drive train early on. Before kickoff, even.

Make a very basic drive before the season, train your drivers on it... and let the programmers work with it. This way, you'll know what you need to do, what improvements are necessary, and also you'll be able to make your drivetrain by the end of the first build week.

When the season does start, immediately design the new one, and build it... all the while you can prototype with the other, and continue to train with it. The earlier everyone gets their hands on your completed drivetrain, the better off you'll be.

[AdamHeard]

Eric and Lavery; Calm Down.

You both know what you're talking about, don't try to nit pick each others mistakes; overall, you're both right.

Also, there is a lot of misinformation in this thread, unless you know what you are talking about and have experience, please refrain from merely speculating.

In terms of straight omni-wheel vs. mecanum each have their pro's and cons. I have seen more succesful mecanum drives than omni wheeled ones, but that's because most of the omni wheeled bots I have seen have been poorly done.

[dtengineering]

We built a mecanum this year... during the six week build period with only a VEX omnidrive holonomic as a practice chassis. It used the banebots encoders to do PID speed control to each wheel. It worked pretty well (thanks to copious amounts of advice available through CD and the FIRST workshop presentation archives), although by the time we had the drive worked out and the robot built there were approximately five minutes of practice for the drivers before we had to drop it into the shipping crate. A typical driver will need much more practice time to maximize the benefits of an omni-drive vs. the benefits of a tank drive. (It is far easier to figure out how to get INTO a pushing battle than to get AROUND one. If you've got an omni drive and you are in a pushing battle... you're probably losing.)

We would have probably won more matches with a tank drive and practice time than with the mecanum... but the mecanums provided a programming challenge for our programming team... the best programmers we have had... or are likely to have... for a few years. They also have inspired the "wow" factor amongst people who view the robot, and generated some interest in the community... and this is where I'll weigh in with two advantages of mecanum over omni...

Omni wheels are cool... but mecanum wheels are COOL!! They just look technologically menacing, and really make people stop to think about how they work. Mecanums are the FIRST equivalent of "spinners" or whatever the flashy shiny rim of the month is in cars these days.

The Andy Mark wheels have the ability to adjust tension in the rollers by tightening the roller screws... in Portland we almost lost an encoder when we fell off a ramp (mecanums can climb, but not like a tank). If we had lost the feedback we could have driven omni in open loop, OR simply tightened down the rollers on a pair of wheels so that they would not roll at all, which would have basically given us the equivalent of a tank drive... we could have easily reset the programming and been back on the field in 30-45 minutes. It would have cost us our omni maneuverability, but we would have still had control. Had EVERYTHING gone screwy with the mecanums, we could have dropped traction wheels on the front and omnis on the back and been a tank drive a few weeks later in Toronto. That is one unmentioned (so far) benefit of having your wheels aligned in a conventional manner. I think the mecanums offer a bit more redundancy and options in the event of a serious competition-season "oops."

Either choice will be a good one to pursue. Keep in mind that (IMHO) it is more difficult to build a competition-winning omni-drive robot (mecanum or omni wheels) than to build a winning tank drive robot. You will end up devoting time and resources to the drive that could otherwise go to programming, actuators, and practicing. That said, when you see your team finally get the drive figured out and go drifting past a defender and slam home a tube... or when your grade 12's get to explain the vector mechanics of the mecanum to an experienced adult engineer who has "never seen a wheel like that before"... that is something you won't get from a tank drive.

Jason

[JesseK]

Originally Posted by dtengineering

We built a mecanum this year... during the six week build period with only a VEX omnidrive holonomic as a practice chassis.

...

You will end up devoting time and resources to the drive that could otherwise go to programming, actuators, and practicing. That said, when you see your team finally get the drive figured out and go drifting past a defender and slam home a tube... or when your grade 12's get to explain the vector mechanics of the mecanum to an experienced adult engineer who has "never seen a wheel like that before"... that is something you won't get from a tank drive.

Jason

This man gets it. It's not about the robots.

I think the point he makes is that even if it sacrifices time and competition success, you have to get your hands dirty or you'll never know for sure.

It is my personal opinion based on what I've seen that mechanum's are better due to traction -- and traction has alot to do with how well your bot drives.

To reduce weight on the mechanum wheel's, you could make custom wheels such as 357's Jester Drive.

Another technical piece that I'd like to know is what kind of power transfer systems do you put on each drive wheel? In Atlanta I saw robots with straight 1:1 drives, 1:5 ratios (or similar) and there was one team that had the 3-speed drill transmissions attached to each mechanum motor, and the Jester Drive CAD shows something like 1:3 connected to 1:3 again.

Also, generally I've seen large-diameter mechanum wheels (are the AM's 6"?), but what is gained or sacrificed in the power and control realm by going to smaller wheels?

[tseres]

ya....for some reason i'm liking the mecanums just because they have more traction. plus, if something went wrong, you could use regular driving code and put regular wheels on the robot....

[T3_1565]

I can't see them giving you much more traction. Omni and mecanums are fairly close when they come to lack or traction. mecanums may have a slight advantage, but not as much as you would think. As I pointed out at our meeting today, mecanums slow down a lot when strafing, which is one of the drawbacks, it won't be uniform no matter what direction you choose. Keep it in mind!!!