The advantage of holonomic drivetrains

[KrazyRoadster]

It seemed to me that it was difficult for robots with holonomic drive trains to be succesful this year. Though these robots may have been more maneuvarable, the loss of traction often greatly hindered their robot. For example, I have seen a robot at sac and and svr which had an excellent shooter and ballpickup mechanism, but its omniwheels made it too easy to defend and psuh around. While I think these drivetrains are as cool as the next guy, I felt that these designs were at a disadvantage inthis years game.

My question is this: Has anyone seen a robot effectively use its holonomic drivetrain to an advantage? If so do you have any videos or pictures? What drive system did they use? ie mechanum, omniwheel, swivel, kiwi And to the teams which decided to use such drive train systems, why did you choose to use them, other than the engineering experience and "cool" factor?

please note that this post is not intended to be offensive to any teams which have used such drive trains. I am interested in these drive trains myself, and I was wondering what was some of the thinking behind using these drives in this years games.

[Andrew Schuetze]

Originally Posted by KrazyRoadster

It seemed to me that it was difficult for robots with holonomic drive trains to be succesful this year.
snip
My question is this: Has anyone seen a robot effectively use its holonomic drivetrain to an advantage? If so do you have any videos or pictures? What drive system did they use? ie mechanum, omniwheel, swivel, kiwi And to the teams which decided to use such drive train systems, why did you choose to use them, other than the engineering experience and "cool" factor?

please note that this post is not intended to be offensive to any teams which have used such drive trains. I am interested in these drive trains myself, and I was wondering what was some of the thinking behind using these drives in this years games.

No offense taken here as we built an omni-wheel system this year looking for manuverability knowing that we may pay the price with defense and ramp climbing.

We play this weekend at LSR, so the proof will be there for us and then at Atlanta. I'd like to dialogue with any all teams using a holonomic drive this year as we do our own assesment. Eventually, we'd like to crab

[Andrew Schuetze]

Originally Posted by Andrew Schuetze

No offense taken here as we built an omni-wheel system this year looking for manuverability knowing that we may pay the price with defense and ramp climbing.

We play this weekend at LSR, so the proof will be there for us and then at Atlanta. I'd like to dialogue with any all teams using a holonomic drive this year as we do our own assesment. Eventually, we'd like to crab

Off topic but somewhat related. How many teams did you see at your two regionals that used a traction wheel set up in the rear with casters in the front? I would say that this set-up has just as much suseptability to a defensive robot yet not gain as much in the area of manuverability. ...

[Eldarion]

Speaking from experience at the Wisonsin regional, a holonomic was definitely at a disadvantage this year. We were pushed and shoved around, and on top of that our drive failed so that we couldn't even drive straight anymore. (We shot at the opponent's goal in autonomous once after executing a neat ground-loop )

Looks like the 4 and 6-wheel drives won out this year, even though holonomic's sooo much cooler.

[Starke]

Originally Posted by Eldarion

Speaking from experience at the Wisonsin regional, a holonomic was definitely at a disadvantage this year. We were pushed and shoved around, and on top of that our drive failed so that we couldn't even drive straight anymore. (We shot at the opponent's goal in autonomous once after executing a neat ground-loop )

Looks like the 4 and 6-wheel drives won out this year, even though holonomic's sooo much cooler.

i have also seen this at many regionals. a robot with holonomic drive would have to be so fast that a 4wd or 6wd bot could not stop you. think of Hot team 67 last year (see below). it was similar drive, i witnessed the robot do a pick and roll maneuver that you see in basketball with. you need to have holonomic drive perfectly right to have it work effectively.

[John Gutmann]

Though I think using a holonomic drive, if used efficiently, you an have a great advantage to the opponent. But in the case of this years game I think a team would just get pushed around too much. Unless of course one could find omni wheels with some traction on the rollers.

[Tim Baird]

Originally Posted by sparksandtabs

Though I think using a holonomic drive, if used efficiently, you an have a great advantage to the opponent. But in the case of this years game I think a team would just get pushed around too much. Unless of course one could find omni wheels with some traction on the rollers.

I agree. Omni wheels would be fantastic this year to zip around the field and collect balls rapidly, but, once you have to stop/slow-down to deliver the balls to the center goal, you're at a huge disadvantage. I've used them in the past on a different team in 2001 when no pushing was involved, and with great success. Perhaps next year...

[Dan Petrovic]

Originally Posted by Starke340

i have also seen this at many regionals. a robot with holonomic drive would have to be so fast that a 4wd or 6wd bot could not stop you. think of Hot team 67 last year (see below). it was similar drive, i witnessed the robot do a pick and roll maneuver that you see in basketball with. you need to have holonomic drive perfectly right to have it work effectively.

Last year the omni-directional drives worked considering the tight space you have to manoeuvre around. There wasn't nearly as much pushing and shoving and there weren't any 3D elements that you had to climb.

[Jeremiah Johnson]

Oh... Witness the BEAST! This was one amazing robot with a crab system. It took quite abit to push them considering they had a holomonic drive. Sure they were pushable but the advantages of their system was greater than the disadvantages.

[Ianworld]

Many people here are getting confused between holonomic and swerve drives. A holonomic drive has 4 omni wheels at the corner of the robot rotated 90 degrees from each other. This allows the robot to effectively move in any direction and spin in place by powering each of the wheels independently a different amount. The reason why people have said that they are weak and easy to push around is because you can never have more than 50% of your power in any one direction. Along similar lines if the wheels are not being driven the robot can be slid across the ground very easily.

A mechanum drive is very similar to a holonomic drive in that it effectively works the same way. It uses complicated mechanum wheels which allow a robot to place its omni-wheels in a front to back configuration like a regular wheeled robot. It however doesn't pose any advantages other than the easier mounting I believe.

A swerve drive is a very different animal which is what the picture of the HOT bot from last year and the BEAST from this year use. They usually have 4 wheels which are rotatable. Usually driven by a motor the wheels are able to rotate in place allowing the robot to drive in any direction. The advantage of this is that you don't sacrifice much traction while gaining a lot of maneuverability. The downside comes in the form of extra motors, extra hardware and a lot of complexity.

As far as I've seen the competition has been dominated 6 wheel drive set ups and treads. Treads and swerve drives are definitely coming out as the top contenders. Treads for their dominance in terms of strength and immovability and swerve drives for being both maneuverable and strong. However teams with just 6" traction wheels from IFI are going to be doing just great this year.

As for the holonomic drives I've seen, they have all broken down under the stress of the competition. They're tough to make really sturdy and powerful and just don't seem up to snuff this year. Its too bad since they're really starting to become common and a bad year like this will slow their adoption I think.

[Eldarion]

Originally Posted by Ianworld

As for the holonomic drives I've seen, they have all broken down under the stress of the competition. They're tough to make really sturdy and powerful and just don't seem up to snuff this year. Its too bad since they're really starting to become common and a bad year like this will slow their adoption I think.

Yeah, I don't think our team will be doing one next year unless the game really demands it. Oh well, lesson learned.

[Dan Petrovic]

Originally Posted by Eldarion

Yeah, I don't think our team will be doing one next year unless the game really demands it. Oh well, lesson learned.

I really want the game to demand it cause that's something I want to do. We'd have to sacrifice our awesome 2-speeds, but still.

I'd much rather have a swerve drive but it's much more complex. We have drawings and the concept for it all done because we planned on doing it for this year. After seeing the game we kind of decided against it.

[irishninja]

Adding to what Ian said this happened with my team this year. We originally were using a tread drive sytem but our tread snapped. By the time we were on our 5th drivetrain and it was Thursday at the NYC regional we swapped for omni wheels in the front, knowing full well we wouldn't be able to get up the ramp.

Consequences of the omni wheels
We were able to turn which let us realign ourselves for shooting very quickly
We were able to be pushed very quickly while shooting
Instead of turning in very small degrees we made huge arcs, although our drivers got that under control
Our omni wheels actually broke
We were able to wreak havoc during the last 20 seconds of the game on the other side stopping from robots getting up the ramp knowing that we couldn't
We usually played backbot

In my opinion the good outweighed the bad, (we came in second), so it was a good tradeoff. If we had been primarily defensive and not played backbot, I'm not so sure the omni wheels would have been a good idea.

[KVermilion]

649 decided on out holonomic drive train, because we had used it last year, and we thought we could step it up, and we did. Keeping the holonomic drive (we call it omnidrive) allowed our mechanical design team to concentrate on our shooter and ball train. If we had not reused our drivetrain scheme from last year, our shooter and ball delivery system would have been likely to suffer. Keeping the holonomic also allowed our programmer and our electronics guy (me) to do some neat stuff, that just wouldn't have happened on a normal drive train.

Simply put, the drive train did not serve us well in quals where one or both of our alliance partners were disabled, or worse, disabled each other. When we had a proper communication and competant alliance partners, it worked well, because we were able to harvest, and line up our shots in a much quicker manner then with a conventional drive train. Of course none of this panned in the QF because were were essentially out there alone, stuck as back bot

[KVermilion]

Originally Posted by Ianworld

As for the holonomic drives I've seen, they have all broken down under the stress of the competition. They're tough to make really sturdy and powerful and just don't seem up to snuff this year. Its too bad since they're really starting to become common and a bad year like this will slow their adoption I think.

649 maintained a working holonomic robot though both SVR and Sac-town. Our holonomic drive gave us minimal issues.