I did a quick search for suction-related topics and didn’t see anything, so here I go.
According to the rules possession is “Controlling the position and movement of a BALL.” G43 allows robots to be in possession of 1 ball as long as G44-46 are not violated. G44 ensures that the ball must be touching the ground. G45 restricts mechanisms above the bumper zone. G46 allows balls to extend 3" into the frame perimeter. I see no restriction then on using suction (or some other holding device) to provide a firm grip on the ball. Thoughts?
The thing I first thought of was suction, but conceivably it is possible to come up with other methods of grabbing the ball with just a three-inch grip. The trident from the field comes to mind, that gets a pretty good grip on the ball with very little contact area.
My only worries would be whether or not the ball comes off the ground at all and whether or not the seams of the ball might cause suction to be lost. Other than that, it seems reasonable.
It would be very hard to implement, because of the seams in the balls. You would need multiple small suction cups, to ensure at least some of them grab. But then they probably wouldn’t be able to hold on if there were any reasonable bumps to the ball.
EDIT: Ah he beat me. And i trident wouldn’t really work because the ball would need to be more than three inches under the bumper to grab the ball in that manner
We came to a similar conclusion, but after looking at the seams of the soccer ball we felt that it would be difficult to maintain suction. Remember the ball must maintain contact with the floor and there will be friction between the ball and the floor trying to pull the ball away. I say run a few tests and see what happens.
I did some tests of this todayt using a vacuum cleaner and a soccer ball. It’s not effective as a mechanism for “aquiring” a ball, becausebut it did seem to do the amount of energy required to have a substantial radius where it sucks enough is enormous, but it did seem to do a .ecent job of holding the ball in place
I have to agree with you hear. We thought about the idea of using suction, as we used it back in 2007. However, the seams of the ball make it very difficult to get good suction. So we abandoned the idea after several attempts of trying to make it work.
<G30-A> allows it to go out to the bumper perimeter for 2 seconds every 2 seconds; <G46> allows a 3-inch penetration into the frame perimeter for the ball. You could do it legally in those constraints.
I think teams might be writing this off to easily. There are many kinds of suction devices which are used in the manufacturing industries. I would like to point you all to this one site which I found doing an easy google search and I am sure there are plenty of other things like this. http://www.anver.com/document/company/vacuum_cups.htm
I am not saying suction is a good solution I just think that there are ways to make it work.
We tried this today. We were at a sponsor’s building, and did not have access to our usual array of building materials. But, we took a mini shop-vac and taped a paper bowl to it as a suction cup using blue painter’s tape. It was easily able to hold the ball upside-down and through some pretty violent shaking. Did the same thing again this afternoon at home with a cut juice container that was skinnier than it was tall, to leave room for a kicking mechanism. It also worked as well.
I think it works well this year because the balls are essentially undeformable. The tubes in 07 deformed a lot, and most teams tried using very small suction cups and very low airflow, thus relying on a near-perfect seal. Team 190 picked up trackballs in 08 with a suction cup. They are more similar to this year’s game piece than the tubes.
The key is to remember physics: First, the more airflow you have, the more leaks you can have. A fan will move much more air than, say, a vacuum pump that uses a piston. Since more air moves faster, the system is able to “fill in” the leaks and maintain a greater negative-pressure on the inside of the suction cup. Second, negative-pressure is pounds per square inch. Surface area varies with the square of the radius of your suction cup, so sucking force varies with the square of your suction cup’s radius. These two things are what allowed our highly primitive device to work.
I suggest everyone try it out before discounting it, it doesn’t take any materials and no more than 10 minutes.
I was having a discussion on the phone with a friend of mine about this (In a more hypothetical tense as to not reveal ideas :P) last night, and she and I determined that it would be a legal option assuming that the ball always is in contact with the ground…
The biggest unknown for me was the stitching of the ball and it’s effect on creating a seal. Now that it sounds like it is possible, I’ll have to do some more digging.
We talked about using air for possibly bringing the balls in from a distance, but with an air grab, you need to make sure the seal is tight or else holding the ball and driving might go to well.
We’re pretty comfortable with the idea of a physical grab, although we haven’t ruled air out, yet.
We found the opposite to be true. Trying to pull the ball in from farther than 1/2" away was futile, but once it gets within about a quarter of an inch, it is immediately sucked on and is very hard to break loose. This is with a mini shop-vac and a paper bowl.
well i believe that it is legal…the only problem i would really have is trying to get it past the bumps if the robot cant fit through the tunnel, isnt it a penalty for a robot picking up a ball?
As long as the ball stays on the ground it is considered POSSESION which is perfectly legal. If the ball comes off the ground that is considered HOLDING which is a penalty.
the key is that any ball on the ground can be in possession of the robot.
Any ball OFF the ground and in possession would be CARRIED… that is a penalty.
Carrying can occur ANYWHERE on the robot… the key is that the ball MUST stay in contact with the ground.
Any vacuum system would have to be designed to keep the ball on the ground…
So a robot that is controlling a ball via suction goes over a bump. At some point, that ball is going to lose contact with the ground. Would this be a carry?
edit: I realize this should be an official Q&A but I’d like to see what you fine folks think.