Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)
-   -   Omni Wheels Survey (http://www.chiefdelphi.com/forums/showthread.php?t=28168)

Ryan F. 25-04-2004 13:38

Re: Omni Wheels
 
Quote:

Originally Posted by bheller
Team 931 used them for the first time this year. We wanted a robot design that was longer than it was wide, for stability with a long arm, and knew that our robot would not turn well with four standard wheels. There are two omnis on the front and two pneumatics on the back.

When we switched to the omnis, it turned too well. Think about it - omnis act as casters when the robot turns. The operator would push the stick to turn and the robot would turn, but it would keep turning when the operator moved the stick back to neutral! The polar moment of inertia kept us going. If you want to go with omnis, I recommend one of three things:
-holonomic drive
-exponential stick control for better turning sensitivity at low speeds
-(best option in my mind) a gyro to counteract the overshoot

I'm trying to get a white paper out on these wheels by midsummer.

Brandon Heller
Mentor Team 931
Alum Team 449

Yeah...you were one of the teams I saw that got me thinking about the omni wheels. We were in the pits right next to you at the MWR...and someone showed them to me.

MOEmaniac 25-04-2004 15:57

Re: Omni Wheels Survey
 
Omni wheels actually worked great for my team this year. But, we didnt have them on the ground the whole time. We had ours attatched to a pnumatic cylinder and when the stoke was out the wheels were down and when the stoke was in they were up. We saw early in the build season that the pnumatic wheels that were provided in the kit didn't allow for smooth turning. So our drive/chassie team came up with the idea of having a set of omni wheels stowed up under the robot and that could be deployed when ever the driver wanted them to. This turned out to be a great asset for out team this year because our driver could easliy maneuver the robot. :D

RogerR 25-04-2004 18:03

Re: Omni Wheels Survey
 
Quote:

Originally Posted by FizMan
... Say you have two motors per set of wheels. Does that mean you can really only apply two motors worth of power in any one direction? So if you were up against a robot that had a tank drive with four motors of power in one direction, you'd only have two to their four?...

this is the main reason that i (personally) don't like "traditional" holonomic setups. mecanum/ilon wheels (such as the ones on these robots) solve this problem to a degree, since it allows you to harness the power of all 4 motors whenever the bot is moving in a direction perpindicular or parrallel to that of wheel rotation. this setup is weakest when it is moving diagonally when only two of the wheels are being powered. in addition, since the rollers overlap, and are rounded, there isn't as much bumping present as there is in the traditional omni-wheel.

David Bridge 25-04-2004 18:28

Re: Omni Wheels Survey
 
For anyone considering the use of omni wheels on their robot, Andy Baker put out a great white paper on how team 45 made theirs and it has all the specs you need to make a great pair yourself. We on team 177 used a slight variation of these wheels on the back of our robot (we kept pneumatic tires on the front) and they allowed for excellent turning. Despite what people say you do not need at least 3 of these wheels by any stretch of the imagination. We recieved many comments from people attending the Philly regional and nationals that they thought we would be very easy to push out of the way with omni wheels and that they were surprised to see us do all of the pushing. If you saw any of our matches you know that defense was a big part of our game, and these magnificent wheels made that possible.
http://www.chiefdelphi.com/forums/pa...le&paperid=226

Alex Cormier 25-04-2004 18:48

Re: Omni Wheels Survey
 
Quote:

Originally Posted by David Bridge
For anyone considering the use of omni wheels on their robot, Andy Baker put out a great white paper on how team 45 made theirs and it has all the specs you need to make a great pair yourself. We on team 177 used a slight variation of these wheels on the back of our robot (we kept pneumatic tires on the front) and they allowed for excellent turning. Despite what people say you do not need at least 3 of these wheels by any stretch of the imagination. We recieved many comments from people attending the Philly regional and nationals that they thought we would be very easy to push out of the way with omni wheels and that they were surprised to see us do all of the pushing. If you saw any of our matches you know that defense was a big part of our game, and these magnificent wheels made that possible.
http://www.chiefdelphi.com/forums/pa...le&paperid=226

i am interested in your omni wheels, and why u guys choose to put them in the back.. please explain. :p

we should get all the robots that have omni wheels and make a big gallery of them, that would look cool.. :cool:

tiffany34990 25-04-2004 19:03

Re: Omni Wheels Survey
 
well it was our first year with omini wheels-- we actually have to fix them up-- only problem i saw was the rubber on the wheels-- a lot of driving just wore them down-- all well-- i'm sure S.P.A.M. won't mind sharign our design-- nobody we saw had a drive system like us-- and we plan on making them better

Rob 25-04-2004 19:06

Re: Omni Wheels Survey
 
I saw a few teams using omni wheels bought from this site:

http://www.omniwheel.com

They have an interesting selection.

Rob

tiffany34990 25-04-2004 19:09

Re: Omni Wheels Survey
 
Quote:

Originally Posted by Rob
I saw a few teams using omni wheels bought from this site:

http://www.omniwheel.com

They have an interesting selection.

Rob


we actually did buy some from that site-- took the whole supply actually for that time period but the wheels didn't work for us--didn't hold our load and no clearance-- i had to actually take down the plastic but that didnt' help much either--so we had to fabricate our own wheels at the end of the 6th weeks-- but we did it-- many teams have seen our wheels they are really cool

David Bridge 25-04-2004 19:17

Re: Omni Wheels Survey
 
Quote:

Originally Posted by Pit Bull 1126
i am interested in your omni wheels, and why u guys choose to put them in the back.. please explain. :p

Well, we put the omni wheels in the back for 2 main reasons... One was that we found they spun better under a larger load (which we had in the back) because we made the neoprene and delrin disks stiff so they wouldn't be free spinning unless there was significant torque from our drive train to make them and with more weight on the wheels they spun better. (Having a lighter load we found caused the robot to bounce when driving straight where as with the heavier load they rotated smoothly) Also it was important for us that the back of the robot was the part doing all the turning since we were a small ball robot and the roller was in the front. Rotating around the front wheels allowed us to stay in front of the balls.

P.J. Baker 25-04-2004 20:01

Re: Omni Wheels Survey
 
Quote:

Originally Posted by Pit Bull 1126
i am interested in your omni wheels, and why u guys choose to put them in the back.. please explain. :p

we should get all the robots that have omni wheels and make a big gallery of them, that would look cool.. :cool:

I'm sure that Dave's answer above is the main reason that the omni wheels are on the back, but there is probably another. 177's robot was meant to have an arm for hanging and grabbing the 2X ball. If the robot had actaully ever been driven while holding a ball out in front, most of the robot's weight would have been over the front wheels. It would have been harder to control and very easy to push sideways if the front wheels had been omni-wheels. This is why our '98 and '00 robots had front wheel drive and rear casters enve though it is generally easier to control a robot with rear wheel drive and front casters (Buzz frequently builds their robots this way when they are picking up balls from the floor).

Max Lobovsky 25-04-2004 23:35

Re: Omni Wheels Survey
 
Quote:

Originally Posted by tiffany34990
well it was our first year with omini wheels-- we actually have to fix them up-- only problem i saw was the rubber on the wheels-- a lot of driving just wore them down-- all well-- i'm sure S.P.A.M. won't mind sharign our design-- nobody we saw had a drive system like us-- and we plan on making them better

i believe 1083 had your same holonomic drive system as nathan pell posted above. I wasn't priveleged enough to see SPAM in action, but i know that 1083 was excellent.

Max Lobovsky 25-04-2004 23:41

Re: Omni Wheels Survey
 
Quote:

Originally Posted by RogerR
this is the main reason that i (personally) don't like "traditional" holonomic setups. mecanum/ilon wheels (such as the ones on these robots) solve this problem to a degree, since it allows you to harness the power of all 4 motors whenever the bot is moving in a direction perpindicular or parrallel to that of wheel rotation. this setup is weakest when it is moving diagonally when only two of the wheels are being powered. in addition, since the rollers overlap, and are rounded, there isn't as much bumping present as there is in the traditional omni-wheel.

I believe you can have all your motors driving in one direction in a 4 wheeled holonomic system.

/-------\
|oooooo|
|oooooo|
|oooooo|
\-------/
for some reason the spaces arent staying between the two walls of my robot so i put o's in the middle

when that robot is moving "up" on the page, all the motors will be applying force in that direction. The one thing i can't seem to figure out is wether 100% of their force is in that direction or 1/sqrt2 of it is in that direction because the wheels are at 45deg angles to the direction we are measuring force in. Now basic physics would say that yeah, its 1/sqrt2 because the force is being applied at a 45deg angle, but the odd thing is, when you get out the other xy component vector, it doesnt seem to be applying a force when the robot drives straight assuming the omniwheels have no significant friction to lateral movement. Now im sure no one got that, but if you did, help me cause im really stumped by this.

RogerR 26-04-2004 02:13

Re: Omni Wheels Survey
 
Quote:

Originally Posted by maxlobovsky
... The one thing i can't seem to figure out is wether 100% of their force is in that direction or 1/sqrt2 of it is in that direction because the wheels are at 45deg angles to the direction we are measuring force in. Now basic physics would say that yeah, its 1/sqrt2 because the force is being applied at a 45deg angle, but the odd thing is, when you get out the other xy component vector, it doesnt seem to be applying a force when the robot drives straight assuming the omniwheels have no significant friction to lateral movement. Now im sure no one got that, but if you did, help me cause im really stumped by this...

the setup you showed is similar to team 180's and 1083's 2004 robots.
i'm pretty confident that that you're right with the second one ,[sqrt2]% of available motor power (but then again, i've been awake way too long). that is the reason i think the mecanum/ilon wheel setup is better, since it allows the motors to apply 100% of their power to full forward/back or left/right movement.

George1902 26-04-2004 02:35

Re: Omni Wheels Survey
 
Quote:

Originally Posted by tiffany34990
nobody we saw had a drive system like us

Tiffany!?

::GASP::

Not only have you guys seen one, but it was at the Florida Regional with you.

George1902 26-04-2004 02:47

Re: Omni Wheels Survey
 
Quote:

Originally Posted by RogerR
the setup you showed is similar to team 180's and 1083's 2004 robots.
i'm pretty confident that that you're right with the second one ,[sqrt2]% of available motor power (but then again, i've been awake way too long). that is the reason i think the mecanum/ilon wheel setup is better, since it allows the motors to apply 100% of their power to full forward/back or left/right movement.

You're so close, Roger. =-]

Both systems get [(sqrt2)/2]% efficiency in the x and y directions. Mecanum wheels have the rollers at 45* angles which produces the same vectors as a normal holonomic platform with its wheels at 45* angles.


All times are GMT -5. The time now is 20:44.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi