Forwards and Backwards Ball Pickup

My team has a design for a “Y” shaped ball holder, launcher, and catcher, like “Draconis.” We thought it would be a good idea to have a robot that could pick up from the front and back, so relaying balls between alliance members is easy, making turning less necessary for assists.

It is well known that an effective rolling ball pickup that has two rollers needs to have the top roller in front of the bottom roller in order to pick up this year’s game ball with few or no problems.

Now, here’s the issue with this: If you have a mechanism that has the top roller always out further than the bottom, flipping it over to pick up from the back will cause the bottom roller to stick out past the top roller, because of the switch between what is considered the “top” and “bottom” roller in your frame of reference, and can make the pick up less effective.

We solved this by having a structure attached to the “Y” that holds another piece of structure that slides inside it, so the length of both arms of the “Y” are variable. We have a design that allows the robot to have the top roller always stick out in front of the bottom, regardless of the “Y” mechanism picking up from the front or back. We did this by having the sliding structure attached to a specially shaped track that allows its length to change as the structure pivots, without using extra power, like motors and pneumatics, or anything complicated like gears or linkages. In fact, there are few moving parts, and it is strictly mechanical, but I won’t let anyone know what it looks like just yet. I’ll post about it later in the season.

Have any other teams considered doing front and back pick up, and have been successful? If so, I would like to know, and have you done something else other than a “Y” shaped mechanism to attempt a front/back pickup? I’m just asking if other people have tried or considered this, since I didn’t find a thread about this, and i’m not asking for anyone to share their designs, but if you wish, feel free to do so.

I don’t think that having a robot pick up from the front and back will make a huge difference. Turning 180 degrees could be just as fast. In my opinion I would rather have one intake that works really well, rather than have a way to pick up from both sides that is just average.


We designed our 2012 robot to pick up from both sides, and it was a useless functionality at best. Turning the robot is way faster than going for the other intake, and the drivers just like to think of one spot on the robot as an “intake zone.” If you saw us play at Davis (and I’m sure you did), you’d have noticed that we had trouble picking up balls, and almost never used our back intake. Moral of the story? Intake speed and ease of alignment are a ton more important than a two sided intake. Save yourself some design complexity and go with a one sided intake.

Maybe the best use of a two-way ball pickup was 624 in 2012. They used it to have an auto mode where they shot balls retrieved from the co-op bridge in addition to their pre-loaded balls

Why not just pick up balls from the back or have a rotating shooter?

not sure…I’m not on 624

I know I was just pointing out that the other options that could have been done and thinking why they would have done it opposed to other ideas.

Overall there is no need for a way to pick up from both sides of a robot this year because it will be just as fast as making a 180 degree turn. Also do to limited space you couldn’t have a shooter that could shoot with a huge distance. But this is just drawing from the ri3d revel videos.

Yes, one intake on each side.


Our team spent a lot of time discussing this today, and reached the conclusion that having a double-sided pickup would cause more problems than it would solve for us. We are using a single roller arm to pick up balls (similar to Build Blitz team JVN), and the double sided intake (via flipping the arm to the other side) resulted in some weird geometry and problems storing the ball. We figured the small advantage gained by not having to turn around in some situations wasn’t really worth the trade off (especially since we’re a rookie team). Instead, we chose to place our pickup on the back of the robot, and the passer/shooter on the front to better facilitate passes.

Although I agree that turning could be just as fast, and if we find that a back/front pickup is not that effective, we can always opt for a simple forwards pickup. But the goal here is to make a front/back pickup that IS just as good as a one sided one.

The only thing I would worry about with turning, is that if a team was to spin around too fast in one place they may risk losing control of the ball due to centripetal force, and the ball may fly off on a tangent, and then again, maybe that could be a way to pass the ball if the drivers know how to do that. If there’s a problem with ball control, obviously the solution is to make a ball holder that eliminates or reduces the risk of the ball falling out.

We like our mechanical arm length switcher idea, because it looks pretty neat so far, and we could possibly win an award for it, but if it proves to be more of a pain in the butt than we expected, and we pass the deadline set for finishing the mechanism prototype, then we move on to a simpler, one sided design. As of now, we find that the design is simple enough to not give us too many hassles.

We’ve had success building some crazier things in the past, like our 2012 robot “Odin” which could stack another robot on top of it to leave more room for triple balances, or make two robot balances easier by stacking, and using one robot to move and balance while the other sat top . It still could hold up to 3 balls, though usually two, was shorter than the bridge when the top doors were down, and when the top doors were open, we could still shoot fairly well, but we ended up being more of a point guard passing balls to teammates. We did get an award for that design, and the front/back pickup is way less complex than a one foot tall robot that stacks a robot on top, but still has a shooter, ball grabber, bridge manipulator, and bumper holder for the robot on top. If you have no idea what I’m talking, about, Team701TV on YouTube has our build season video on it. Here’s a pretty popular video of our first successful unassisted stack in a competition.

Anyway, my point is that I have the confidence in my team to build this system successfully, because it’s definitely not the hardest thing we’ve ever done. My team is all about taking reasonable risks and making a unique robot, and sometimes that just makes the season more fun, challenging, interesting, and especially more fulfilling than it already is when the students succeed and also receive recognition for it.

But do you really need a top and bottom roller?

It all depends on what you’re trying to do. For us, we are trying to grab and lift the ball up with an arm mechanism to launch it, and pick up both sides so two rollers can be useful.

With a robot has a one roller over-the-bumper-intake, or most robot designs that I could think of for that matter, two rollers wouldn’t be as necessary.

I don’t think if you can solve this problem you should take on the complexity of a double sided intake. I know what you’re trying to do and it seems tempting to pickup from both sides but I think once you abandon double sided pickup you’ll find it’s a lot easier to package everything anyway.

To answer your question: I bet a properly sized chain linkage could angle your claw such that it’s pointing “down” on either side of the robot. Doesn’t even need an extra motor if you do it right.

We had a rotating turret, but we stopped using it by our first regional.

The double sided intake was extremely useful, and was one reason we were able to keep up with other top teams, as well as for a 4-ball autonomous and effective bridge tipper, we were able to advance to the semifinals at CMP and Number 2 seed at IRI. But there were a lot of game pieces in 2012, so being able to pickup from multiple sides was a bonus (I believe 1717 had a two-sided pickup as well, and it worked great for them too just as it did for us).

This year, you could make a case for a double sided pickup, but you could just turn around as well. Comes down to what your team wants to do and how you integrate them.

Also…I don’t think I’ve seen many 2-roller intakes in any year? 2011 was the only year I saw some effective 2 roller systems, but 2012 and 2013 all featured 1-roller + ramp intakes.

Double Roller intakes have been around for quite a while. I can remember them in 2007, 2008, 2009, 2010, and 2011. Since 2008 and 2011 were the most prominent, and this game being like 2008, I see a lot of double rollers being utilized by teams.

Lots of discussion about moving intakes around to pick up from multiple sides…what about two intakes? It seems like it might be hard to package everything.

I see, before my time :slight_smile:

I would have to say that if a back/front intake requires an extra motor compared to one that is one sided, it isn’t worth doing. Our design is simple (to my own standards, at least), easy to build, and requires no extra motors or power sources, so we want to give it a shot

The design we have planned takes up as much space as a single sided intake, but has a few extra pieces, of course. It shouldn’t be hard to package at all. I don’t know about how other teams are trying this, if they are, and if their design is difficult to make fit especially if their intakes are separate.

I’ve been working on a similar design concept (a well packaged claw on an arm pivoting about the center of the robot). What ultimately drove me away from using two rollers and thus picking up on both sides wasn’t the extra motor requirement. For this design I wanted the rollers to articulate out of the way to allow for a clean fire from some spring force. With passive forks on the bottom instead of rollers, actuation for this motion became a lot simpler and lighter. If you’re thinking something else entirely, then carry on.