I knew that blue cloth looked familiar, saw you guys compete at regal eagle last sunday
Yikes I can definitely understand how this would happen, my team originally wanted to pick up tipped cones by angling our claw downward like that and scooping. Our wheeled intake prototype was too wide to pick up cones like that but needed to be large enough to get cubes
our team uses a claw with linear motion powered by two motors to let us horizontally move the game piece because no swerve. We have lost game pieces every so often but it hasn’t been too bad
217 2011 as well.
4 Likes
2992 took a distinctive approach this year. They have a claw, but they only pick up game pieces that are already inside the robot, and rotated to the proper orientation. They also have a separate roller intake to get game pieces from the floor into their cuer, but don’t use it very much - most of their game pieces are dropped directly into their cuer from the single SUBSTATION. Definitely the most effective non-roller claw I’ve seen this year.
1 Like
4451 did the same at Anderson the week before. I will say after watching every match at both events (#justMCGAThings), both bots occasionally struggled to get a good grip on their game piece even when it was already inside the robot. Weird edgecase for touch-it-own-it for sure.
@D.Allred Any thoughts on your architecture and if you’d do something different in the future?
While I can’t speak to 4451, from what I saw, 2992’s claw issues were from two sources, one of which appears resolved, and I expect the other will be addressed in the next eight days (that is, before Bayou).
The first was that they were dropping cones and often holding them at unfortunate angles. This was due mostly to the cones hitting the edge of their carousel (or bucket/cuer/spindexer/whatever you want to call it). They fixed this in software by having the claw rise farther before swinging into the placement position.
The second is dealing with cones which land vertically in the carousel - the setup is designed for cones on their side. Often they were able to get these knocked over by doing some fit-and-start driving.
Another thing about their setup that seems easy to improve (from me on the outside, so I may be completely out to lunch) is that other than running the carousel, they don’t seem to get the game piece with the claw until they’re essentially in scoring position. It seems that if they had an operator (separate from the driver) who tended to this as they made the cross-field transit, they could shave another couple of seconds off of their cycle time, but the handful of times I watched the claw during a transit from SUBSTATION to GRID, it wasn’t actively trying to secure the game piece in the carousel.
Our architecture is based on a similar hand-off design develop by one of the Ri3D teams plus 4481. We did this because we were a tank drive and wanted to pick up in the front and score out of the back. All game piece acquisition is floor load in any orientation.
The claw that picks up out of the spindexer is just a pneumatic gripper. We sometimes catch the flange of the cone instead of the body, but in general don’t have too much problem.
We played Anderson with no automation other that Charge Station balance. We now have everything automated except spindexer alignment which is done via camera stream. The pneumatic claw works fine. We were thinking about making it a powered claw but we are up against the height limit already.
We got rid of the standing cone problem by adding a “turbo mode” to the spindexer. (Falcon motor go zoom…) We do try to orient the game piece and grip it before getting into scoring position. It’s basically our whole concept of pick it up quick and orient in transit.
I definitely wouldn’t recommend a handoff architecture for swerve. Keep it simple with rolly wheels. We should be swervin’ by next year just in time for a 2016 redux.
David
2 Likes
It seems like for this game a wheeled over the bumper intake that has a handoff to a conventional claw mechanism is the most efficient.
Def the wheel, you wont have to line it up like a claw, it will get sucked in and will stay sturdy if you do it right - 4336 used an everybot like claw and it worked perfectly with how fast they were.
1 Like
Just as a heed of warning, make sure that you have compression from all sides!!!
This intake worked great for cubes, but for cones, we had no backside compression on the cone and it fell through really easily when we turned or took a hit. Had we added some kind of backstop or made the compression tighter, it would have worked really well, but then we would have sacrificed width and opening which would have lead to less margin of error for the driver. We ended up going with a claw that works super well and clamps both pieces nicely.
Iterations save lives :)))
5 Likes
343 at Anderson had a similar process. Aligning to the station and dropping a cone in then gripping it with a simple claw. From watching matches it appears that the most difficult part of that process was aligning correctly to the substation (would be easier with swerve)
They too have a roller intake, although I believe it only works with cubes (maybe they’ll have cones from the floor figured out by Hartsville)
I concur this seems to be a really good way to improve the effectiveness of a claw.
Don’t let my icon fool you, I don’t have all the insider info for 343, alum and former mentor who moved away.
i think the less the drivers have to do the better, lots and lots of teams take lots of time up on the field using their claw systems, lining up, and then grabing, its a big expense of time esspecially when the alternative is lining up, grabbing and going. claw designs work theoretically but as far as simplifying game piece manipulation and speeding up cycle times the rollers are going to do that best.
I would challenge that an intake that can intake off the floor without needing a handoff at all is most efficient.
5 Likes
We actually saw one that looked like a crane claw…My opinion…it was gangly and not very effective.
Claws with rollers…very effective.
Rollers…very effective.
Claw…IF they had like a paddle to upright cones…equal pressure…very effective.
My team used a claw intake for our first 2 competition but we are going to change it to wheeled intake after observation of other teams using wheeled intake like 2075.
1 Like
This game has been kinder to delicate ended effectors picking game pieces off the floor than I anticipated.
We’ll see if that changes with the developing game meta.
huge kudos to you guys for coming up with that claw, it works amazingly
1 Like