Contested Containers

[JesseK]

Originally Posted by bowmanb

Our bot can already build a stack of 6 from the load station as fast as our human can feed them. We are definitely NOT an elite team, so I see stacking from the chutes as being a lot faster than this discussion is accounting for. If just one of our partners can handle the RC bins (pick them up, get them noodled and place them on a short stack) then we will be able to use all the bins we can get.

I think the more consistent scenario for this RC-centric strategy is the 5-stack, given the 6 stacks of 5 behind the glass and partners who are more likely to be able to stack on a 5 than they are on a 6.

I can say that reliably controlling a RC by only touching its lowest 6" (for the 6-stack) is pretty tough after the RC is in the air. Doable, but tough. I know that our prototypes have been much more reliable if they're closer to the RC's center of mass, which is about 16" off the ground. It's still early though.

[asid61]

Originally Posted by JesseK

I've had to re-iterate this to the veteran mentors on my team, since they're so used to the frame of mind that a low-scoring 'win' can be successful this year. That mindset won't even make it out of QF's.

10 Quals, 2 QF's, 3SF's, 2-3 Finals = 18 matches potentially at the average event.
The competitiveness of the containers isn't what matters for 15 / 18 matches, only the raw score of our own alliance. If the "opponents" want to get those containers to just keep them away from us, let them waste their time. We'll just stack our 3 RC's on a bigger stack with a less cluttered field.

It should really be emphasized that an entire robot should not focus on a strategy that only plays out 16% of the time for only (on average 6 / 50) 12% of teams. The strategy only works before finals if the alliance uses those containers. Given some of the discussions I've seen/heard, there is so little emphasis on what to do after the RC's are acquired, or even where to put them so they're out of the way of stacking.

Getting to the containers on the step in a controlled manner and getting a single container from the step in a controlled manner is much more valuable long-term than something that "wrangles" 4 at a time back to the alliance all willy-nilly.

But you don't need the most points to win; you need only be picked by a high-seeded alliance. If you design both mechanisms, you can swap them out if other teams are better at grabbing the cans than you, and stack normally.
Denying points is important, as ultimately, it comes down to elims.

[MrForbes]

In my limited experience winning regionals, it's helpful to either have the highest scoring robot, or to be indispensable to the team with the highest scoring robot.

We don't look at what it takes to be picked...we look at what it takes to be the team making the first pick.

[bowmanb]

Originally Posted by JesseK

I think the more consistent scenario for this RC-centric strategy is the 5-stack, given the 6 stacks of 5 behind the glass and partners who are more likely to be able to stack on a 5 than they are on a 6.

We are thinking that making two 3-stacks then placing a noodled bin on one and then placing that stack on the second one for 42 points is a viable strategy. We are working toward optimizing tote-stacking and bin handling so that we can do either/both.

[Colin Small]

Originally Posted by asid61

But you don't need the most points to win; you need only be picked by a high-seeded alliance. If you design both mechanisms, you can swap them out if other teams are better at grabbing the cans than you, and stack normally.
Denying points is important, as ultimately, it comes down to elims.

Because not only are you denying the other alliance the container, you're getting the container for your team. Using them is up to you, but you're still denying the alliance members on the other team those points. My reasoning may be wrong, but points the other alliance doesn't score means more points they'd be farther away from you in the standings or by less points they'd pull away from you.

[cglrcng]

I will make a set of prediction(s) now. (I would have/could made this prediction on game release day 1 easily also).

A team that can quickly snatch 2~4 RC's from the step in Auto, time after time (say 80~90% or better), (or 2 in Auto & can get to the Robot Set position, & 2 more immediately when teleop begins, if they are still left there of course after the first 15 seconds...very doubtful at least 25%~50% of the time, as I will also predict now), WILL BE playing on both Einstein fields this year.

In fact....You can be the lowest actual scoring personal single robot on your Alliance, in qualifying, round to round, every single time or event. BUT, if you can guarantee at least 2 or more RC's off the shelf in Auto onto your side of the field, and more, on a consistant basis, and still stack (or even push around stacks of), a few Totes, cap a few Stacks w/ litter loaded RC's, upright RC's that have fallen over (~THIS is really important also), handle the RC's to get the litter in them at the human station, your robot will be "worth its weight in gold" to ANY alliance in the Playoffs or at Champs IF teams are scouting properly. (Great INBOUNDING STRAIGHT THROUGH ROBOTS last year were worth their weight in gold also, though they really didn't do that much as far as scoring points personally themselves...They were ALLIANCE COMPLEMENTARY, and very necessary to score & win). They helped the High Scorer super shooter, shoot faster and gain more points.

Another prediction here, they (THE SUPER Shelf RC Capture Bots), will also score in the top 1/3 of all QPA's more often than not also. Denying possible points to the opposition (and allowing your alliance the opportunity to use any and all of said stolen/snatched RC's as "HUGE points multiplyers" on any stack of grey totes from 1~6 on a scoring platform will often mean more than just the match)...It will always equal YOUR ALLIANCE more/higher points possibillities, while also denying higher possible points to the opposition ALLIANCE. (Doublemint Gum!)

I will also predict, that HUGE AUTO and Tele-op tug of wars will no doubt happen often, and RC Snatchers will go through more than a few actual repairs before the season is over...The more robust the Robot Shelf RC Snatchers are, the better they will fare. The RC Lids...Not so much. (I would be ordering more than a few extra lids for those RC's if I were preparing fields for FIRST!) They can (And Should), be ordered separately.

Storage of the RC's once captured 2 X 2 preferably, is rather easy, then grab them one by one, take to human player station, get loaded w/ litter, drop on any 1 grey tote or higher stack on the platform(s) (that would score it no matter what), any elevator or forklift loader can lift the totes and Litter loaded RC onto a higher stack later. Early on until stacks start, they could also be loaded and stored right in front of the player station glass momentarily only in the middle (careful there though, you wouldn't want to waste those very important points multipliers).

Last prediction....The Build for Playoffs/Champs/Any Q round "SUPER SHELF RC Snatcher" (Will not be in the slightest ever overlooked, they will be up front and on display just as much as any 3 Yellow Tote Auto Stacker, or 4 Tote Shelf Stacker, though they won't need to be as accurate, ever...As, on your side of the shelf/field, is all that is necessary (In the Auto Zone BETTER). But, also clean up the messes you make too!)....They also Would "NOT BE" A 3rd. pick, they would be a second pick, and possibly an Alliance Captain & a picker.

If your Alliance very often has 5~7 of those RC's avail. on your side to use, you will use them more, you will score higher more often...You will be in the top 10~15 anywhere after 10 Q rounds...Almost anyone playing this game this year will be able to stack Totes & drive I certainly hope.

The much more rare (and almost as equally important), utility bot that I will also predict, will be the one that can lift & rotate / or right a Tote that is not rightside up (those upside down nasty ones will be the very worst, and IT will happen more than just what is already on the field to begin with I think).

Now, the real moral question becomes (for those daring to choose to go that route), when faced w/ a null / or extra Q round (that rare event where you cannot gain qualifying average points, but do still play a Q match). Do you use that Shelf RC steal that round, or not? (It is a rare time when helping your actual Alliance partners to gain a higher QPA, can possibly actually hurt your own teams actual Q position). Does not seem right though does it?

WELCOME to a nuance of no more Win/Loss/Tie, & the new QP Average scoring dilemma.
_____________________________
Lastly....If you have not done so already. Label all your RC's/Totes/ & Noodles that you may be planning on taking to competitions / practices, etc. w/ your Team #'s immediately. (I can see many taking them to practice fields etc., and losing them very easily). If they are pre-marked now, it will be much easier to determine your personal ownership later. (It would sure be nasty to load those up those practice game pcs. w/ stuff and take them into the venues, then not have those storage containers when you went to pack up again at the end of the event). LOL.

[cglrcng]

Were I building personally a "Super Shelf RC Snatching Bot" it would be a utilitarian Tote Stacker, RC Handler/RC righter/tall capper (I'd be asking my Alliance partners to work together if possible to build a 4 High Tote wall on the far/left platform first using everything they can reach from the landfill first and the right Human Station (please allow me to work from the left human station as I will store RC's along the left wall and center glass). Please lock them (Totes no more than 4 high), in tight together, and push them tight into the backstop as a wall~This will be key to later capped wall stability!

(I will concentrate on nothing but snatching 2~4 RC's, then righting all of them, loading them with litter, then stacking an RC on 2 Totes and capping the stack~One on every other 6 high stack from the left first/far platform, 2+RC+Litter, 2, 2+RC+Litter, 2, 2+RC+Litter, Etc. (once the far platform is finished ~THE FAR LEFT MONEY PLATFORM THAT IS....NOBODY dares goes near it please! 5 RC's capped easily fit a tight stacked wall, so 10 Totes Wide X 6 Tall=60 Totes+5 RC's+5 Litters)...While they build out the near platform wall last 4 high from the landfill and the other human station backstop first again (one should also push all the upside down totes to the right side of the field early if you cannot right them at least, before building that left wall too much, so they can be used as the lower layer pushed into place if necessary, on the right near platform as the very bottom layer). The RC Snatch Bot (or whoever is actually closest and fastest), should deliver the Yellow Totes to the Shelf and cap/stack co-op if possible.

I work on using up the balance of those RC's capping the remaining stacks on the close right wall. (Key is to get the multiplyers on the walls ASAP on top of at least 1 or more totes).

If there is still time left on the closer right platform, one bot can lift from one side and allow others to carefully slide totes underneath your shorter capped stacks, then lower carefully, and all back away, again carefully before that nasty buzzer sounds). Alliance Coach giving a LOUD 5 second countdown.

Risking moving around those possible remaining floor litter after 2 walls are built is points/pound foolish if it could possibly take down the walls. (Minus 4 for plus 1, is nothing compared to knocking down huge stacks of points!)

In the end capping 7 stacks (5 FL +2 CR Platforms), is not out of the question. Empty the Landfill first w/ 2~3 bots, then allowing the RC Bot to use the closest HP Station. Then down to only the left HP Station. Then once that left/far platform is finished (one bot in right side of landfill, 1 at right station as RC bot finishes up capping using left HP station only.

Well, there is a strategy theory anyway.

[cglrcng]

Originally Posted by wilderbuchanan

Do you guys really think that that many Totes will be stacked. I mean the 30 at the feeder station plus 28 in the landfill. Thats 60 totes by itself. The time spent stacking all of those could be spent on stacking up to 7 recycling bins resulting in many more point. Also throw in the coopertition and i think that stacking 60 totes, unless a team develops some huge machine that can stack a tote a second etc, will be unlikely at most of the regional events. I do think that going in getting the recycling bins in autonomous could be very beneficial.

There is a maximum possibillity of up to 40 Totes for each (one side steals or captures the whole shelf, though rarely will that occur I would think), side in the landfill and 30 outside of the field (=70 possible total Grey Totes for 1 side, which would leave the other only 58 Tote Max., or some combination thereof) + a maximum of 7 RC's at the beginning of the game.

(Stealing or Capturing from the shelf all of both RC's and all the Grey Totes), is the smartest thing to attempt to do. 3 Robots building 6 high capped stacks will be fairly fast as many can stack locally in different areas, and deliver far more than 1 tote per trip. (One consistant tightly packed scoring platform wall 6 High w/ every other 6 High stack capped w/ an RC =at least 60 Totes+5 RC's+5 Litter for just 1 wall. There is another scoring platform to fill yet.

You just did not account for the other 12 Grey contested Totes also on the shelf, in your formula above. As more and more practice progresses, stacking will become second nature. And working together will become first nature. Stack, deliver stacks, someone backs into and packs the wall towards the backstop, cap the stack. Off to stack more. Tight locked in walls will be the big very stable winners!

There will be plenty of bots that can stack 3~5 high w/ an RC cap and deliver them to cap the shorter built tight packed stacks. A 1~2 high Tote Stack w/ an RC+Litter cap is very stable. Easy to cap an existing 4~5 high tight packed Tote stack on the platform without having to grapple that last
3" of the bottom of the RC and lift into place last. Lifting a short stack without driving anywhere will allow others from the other side to bottom feed in easily. There are many ways to work together to build those walls. Those that do work together well, will be rewarded highly.

Building a tight wall will require you to cap every other stack maximum though, not every stack, as the RC's are much wider at the tops than the bottoms.

[Jarridj4009]

Originally Posted by JesseK

I think the more consistent scenario for this RC-centric strategy is the 5-stack, given the 6 stacks of 5 behind the glass and partners who are more likely to be able to stack on a 5 than they are on a 6.

I can say that reliably controlling a RC by only touching its lowest 6" (for the 6-stack) is pretty tough after the RC is in the air. Doable, but tough. I know that our prototypes have been much more reliable if they're closer to the RC's center of mass, which is about 16" off the ground. It's still early though.

Our robot still needs a lot of upgrades to speed, accuracy, and grip ability, but so far we have successfully stacked a RC on a stack of 6 crates fairly easily even without any automated levels. I think teams that planned to do this from the start won't have any problems.

[JesseK]

Originally Posted by Jarridj4009

Our robot still needs a lot of upgrades to speed, accuracy, and grip ability, but so far we have successfully stacked a RC on a stack of 6 crates fairly easily even without any automated levels. I think teams that planned to do this from the start won't have any problems.

Video speaks louder than words . Really I'm impressed though; it was a difficult thing to do, and takes a bit of specialization to pull off. More seriously, you may want to follow this Q&A question. I only thought of it when perusing the other Q&A's, then wondered if our robot was over 76" under certain circumstances. Turns out we're ok regardless of the ruling, but I get the impression that other teams may not be.

In order to maintain compliance with [G22], and to get a RC onto a 6-stack (~70.1" floor-top) on the floor, the robot must maintain grip on the bottom 8" of the RC. If the robot remains on the floor and the 6-stack is on the platform, the robot must maintain grip on the bottom 6" of the RC. Depending on the answer to the Q&A above, and depending on the type/geometry of the lift, if the robot ever intends to drive over the platforms each of these two numbers may be decreased by about 2".

[pntbll1313]

Originally Posted by JesseK

Video speaks louder than words . Really I'm impressed though; it was a difficult thing to do, and takes a bit of specialization to pull off. More seriously, you may want to follow this Q&A question.

Doesn't the blue box under G22 specifically say that it is OK to be higher than 6'6" from the floor when on the scoring platform? That's how I've been reading it.

For reference, the height restriction matches the height of the ALLIANCE WALL. However, it is expected that ROBOT that are on a SCORING PLATFORM may actually be slightly taller than the ALLIANCE WALL due to the elevation of the SCORING PLATFORM. This is permitted and not a violation of G22.
An example of a strategic violation of G22 would be intentionally extending above the height restriction in an effort to complete a scoring action.