pic: 2015, Year of swerves?

[mrnoble]

Originally Posted by asid61

Well, "swerve is never necessary" is like "moving is never necessary". You can sit still for a whole match with blinking lights and compete with that. Great. However, it's probably advantageous to move a little bit at least.

Ask 254 (from whom the quote is sourced) if not driving swerve is like not driving.

[hzheng_449]

Originally Posted by Thad House

with all the COTS options I could see alot of teams doing it.

While the COTS swerve modules certainly have made *machining* a swerve as simple (if not simpler) than machining a WCD, the mechanism is only part of swerve.

In order to get the full benefits of swerve (or really any omni direction drive for that matter), you really need to be able to implement a good control scheme that is exponentially more difficult to do than a control scheme for WCD.

Also as a side note: I think that some of the initially negative posts about swerve on this thread we're saying "Swerve is useless", but were saying "A lot of teams are going to underestimate the challenge of implementing swerve during build season and will subsequently not have functional/effective drive bases come competition time". (Disclaimer: That's just my interpretation and I don't want to put words into anyone's mouth)

[Abhishek R]

Originally Posted by asid61

I thought that graph was pretty accurate.
Anyway, many swerves may not be built for 2015. 2016 will have more swerve drives IMO.

Well, "swerve is never necessary" is like "moving is never necessary". You can sit still for a whole match with blinking lights and compete with that. Great. However, it's probably advantageous to move a little bit at least.

That argument doesn't make a whole lot of sense. I can think of several counterexamples, but I would rather you reword the post because I think you might have meant something else...I assure you as a driver myself that swerve is never necessary, whereas moving is. The best drivetrain is the one that gets you from point A to B the quickest, and if that drivetrain for you is swerve, then that works out nice. But for a lot of teams, it seems like a standard tank drive accomplishes that goal just fine.

[MichaelBick]

Originally Posted by Abhishek R

That argument doesn't make a whole lot of sense. I can think of several counterexamples, but I would rather you reword the post because I think you might have meant something else...I assure you as a driver myself that swerve is never necessary, whereas moving is. The best drivetrain is the one that gets you from point A to B the quickest, and if that drivetrain for you is swerve, then that works out nice. But for a lot of teams, it seems like a standard tank drive accomplishes that goal just fine.

The goals of a good drivetrain depend on the team. I think for most teams a reliable and easily maintainable drivetrain are far more important than everything else. For many teams, agility and speed are far lower on that list than you would think, and therefore swerve isn't prioritized.

[Abhishek R]

Originally Posted by MichaelBick

The goals of a good drivetrain depend on the team. I think for most teams a reliable and easily maintainable drivetrain are far more important than everything else. For many teams, agility and speed are far lower on that list than you would think, and therefore swerve isn't prioritized.

That's true. I think what I meant was that you can still achieve a high level of speed and agility without attempting the complex nature of swerve and still have a championship level robot.

[efoote868]

Originally Posted by asid61

I thought that graph was pretty accurate.
Well, "swerve is never necessary" is like "moving is never necessary". You can sit still for a whole match with blinking lights and compete with that. Great. However, it's probably advantageous to move a little bit at least.

I'm going to agree wholeheartedly with that last sentence, which is why I would strongly recommend to any team NOT to use swerve. I've witnessed veteran teams (including my former team) sit still for entire matches because of problems exclusive to swerve.

[jman4747]

I don't know why you all are here but I joind FRC to make cool robots and because when I saw 2010 champs back when I was too young to compete I thought that there was no way I could do all that (looking at some team electrical system. I could design a WCD in a day or two now.

I don't care about WCD. I don't want to try and shave a 1/4lb of my sixth iteration of a WCD. I want to actually look around in the Math section in LabVIEW and know what an "arc tangent" is. Enough with this 6WD 8WD is on some special peice of carpet. I hope I never win champs with the same old boring robot as 2K other teams. It better be the coolest most difficult, fun to talk about and look at and think about and know that it was hard and I learned somthing.

If you don't have money figure out how to do it with less.
If you don't have the tools figure out how make it without.
Be an engineer solve the challenge.

[Woolly]

Originally Posted by Michael Hill

This will be the return of 1/4 of the robots not moving at their first competition if it is.

If my team ever wanted to try swerve again, I would be pretty adamant that we would need to have the ability to put 4 brackets over the swerve modules that lock them straight forward, and have code ready to go so that a toggle on the driver station puts the robot in tank mode. Swerve is cool, but considering all the things that need to be perfect for it to be preferable to a tank drive kind of requires having a quick way to fall back if something is amiss.

[magnets]

Originally Posted by jman4747

I hope I never win champs with the same old boring robot as 2K other teams.

I have yet to see a single team be on the winning alliance at champs with a "boring" robot. Myself and many others agree that if your team doesn't try its hardest to build a competitive robot, your team is doing a disservice to the teams that are paired with you.

It's your responsibility that your entry to the FIRST Robotics Competition be your team's best competitive entry. If you decide that swerve is the best thing for your team, and it turns out to be unreliable, I might think you made a silly, but understandable, mistake. If you realize that swerve will lower both your rankings, and the rankings of those who must play with you, and you still decide to build it, it reflects poorly on your team.

[Abhishek R]

Originally Posted by jman4747

I don't know why you all are here but I joind FRC to make cool robots and because when I saw 2010 champs back when I was too young to compete I thought that there was no way I could do all that (looking at some team electrical system. I could design a WCD in a day or two now.

I don't care about WCD. I don't want to try and shave a 1/4lb of my sixth iteration of a WCD. I want to actually look around in the Math section in LabVIEW and know what an "arc tangent" is. Enough with this 6WD 8WD is on some special peice of carpet. I hope I never win champs with the same old boring robot as 2K other teams. It better be the coolest most difficult, fun to talk about and look at and think about and know that it was hard and I learned somthing.

If you don't have money figure out how to do it with less.
If you don't have the tools figure out how make it without.
Be an engineer solve the challenge.

I think there's a fundamental difference here. I can tell you that any team that has won champs did not win with the "same boring robot as 2000 other teams." They also worked hard on something they thought was cool, and I'm willing to bet it was difficult for them to build that level of robot as well.

Besides that, you can definitely be unique in other ways than the drivetrain. 254 this year was pretty much one of a kind despite using a fairly standard drivetrain (though they were the one of the first teams to popularize the system).

[asid61]

Originally Posted by Abhishek R

That argument doesn't make a whole lot of sense. I can think of several counterexamples, but I would rather you reword the post because I think you might have meant something else...I assure you as a driver myself that swerve is never necessary, whereas moving is. The best drivetrain is the one that gets you from point A to B the quickest, and if that drivetrain for you is swerve, then that works out nice. But for a lot of teams, it seems like a standard tank drive accomplishes that goal just fine.

Well, I look at it this way:
When somebody says, "Swerve is never necessary" it's basically just stating a universal truth. That quote just kind of irks me because of that. Yes, swerve is never necessary. Neither is driving. But, driving, and swerve drives, are advantageous. And it's that distinction that the quote totally misses.

What I mean is that swerve, when practiced with in the offseason and with a good driver, is almost always a benefit IMO. You're getting all the capabilities of a tank, it can be configured to drive like a tank, and if you want to you can switch over to swerve anytime. Now, if you just try to make a swerve when the season begins, I can see how a team might flop. But provided you actually prototype in the offseason, it's a lot less likely to fail.
Open-source the designs and code, manufacture early in the season, and you're good. A well-designed swerve drive is lightweight and compact nowadays, to the point where it can compete with a WCD. For some teams it's not easy to manufacture it at all, and for them swerve is in fact disadvantageous. But if you can get it made fast and you have working code pre-season, I don't see the inherent disadvantages of a swerve drive.

"The best drivetrain" can mean a lot of things depending on the game. But provided the programming and design work are done pre-season, swerve is not inherently disadvantageous.

[AdamHeard]

Have your fielded a swerve drive during season?

Originally Posted by asid61

Well, I look at it this way:
When somebody says, "Swerve is never necessary" it's basically just stating a universal truth. That quote just kind of irks me because of that. Yes, swerve is never necessary. Neither is driving. But, driving, and swerve drives, are advantageous. And it's that distinction that the quote totally misses.

What I mean is that swerve, when practiced with in the offseason and with a good driver, is almost always a benefit IMO. You're getting all the capabilities of a tank, it can be configured to drive like a tank, and if you want to you can switch over to swerve anytime. Now, if you just try to make a swerve when the season begins, I can see how a team might flop. But provided you actually prototype in the offseason, it's a lot less likely to fail.
Open-source the designs and code, manufacture early in the season, and you're good. A well-designed swerve drive is lightweight and compact nowadays, to the point where it can compete with a WCD. For some teams it's not easy to manufacture it at all, and for them swerve is in fact disadvantageous. But if you can get it made fast and you have working code pre-season, I don't see the inherent disadvantages of a swerve drive.

"The best drivetrain" can mean a lot of things depending on the game. But provided the programming and design work are done pre-season, swerve is not inherently disadvantageous.

[jman4747]

Originally Posted by magnets

It's your responsibility that your entry to the FIRST Robotics Competition be your team's best competitive entry. If you decide that swerve is the best thing for your team, and it turns out to be unreliable, I might think you made a silly, but understandable, mistake. If you realize that swerve will lower both your rankings, and the rankings of those who must play with you, and you still decide to build it, it reflects poorly on your team.

I think I have a responsibility to be the best I can and for this team to be the best it can this year and next. To me that means that I should try to push the limits of my current abilities. I realize I sound too condicending towards teams who did make it to and win Einstein, however winning is not the primary perpose of building a great machine, building a great machine is. There is a difference between doing your best in rankings/placement than doing your best as an engineer and student. Our "best" issn't what gives us the best chance of winning it's what was hardest for us regardless of necessity.

Also "same old robot as 2k other teams" is unfair, I'm sorry. But if I can help it I won't try to have my systems fundamentally similar to most teams because that's the easiest rout. I'll make it because it's the best I can do with what I have.

[asid61]

Originally Posted by AdamHeard

Have your fielded a swerve drive during season?

Fair point. You can only get so much by observing.
Not yet, much to my annoyance. However, if I didn't hold the belief that I outlined in my posts, then we never would.
Could you explain how a design that works in the offseason, code and mechanism, would fail in the season?

[artK]

Originally Posted by jman4747

I don't care about WCD. I don't want to try and shave a 1/4lb of my sixth iteration of a WCD.

Iteration isn't always about saving weight, even on a drive base. In my four years with 254, we never used the same drivetrain twice. In 2012, we attached pistons and small levers to help push us over the bump. In 2013, we developed a PTO in the gearbox to climb the pyramid. In 2014, we had to make the gearbox have a low profile so the balls could fit inside the robot.

I want to actually look around in the Math section in LabVIEW and know what an "arc tangent" is.

Personally, I would want to know what quintic hermite spline interpolation is, but I always liked polynomials more than trigonometry.

Enough with this 6WD 8WD is on some special piece of carpet. I hope I never win champs with the same old boring robot as 2K other teams. It better be the coolest most difficult, fun to talk about and look at and think about and know that it was hard and I learned something.

As a developer of a boring robot, I feel so awful about winning the championship, because our robot was so boring and uninspiring to watch.