Two Boulder Auto

[cglrcng]

Originally Posted by hardcopi

I don't think the accuracy will be the downfall. Accuracy isn't beyond any of our teams. ie: https://www.youtube.com/watch?v=N995FopDHgU

I do wonder how many fouls will be racked up before the 2 boulder auto will be dialed in. Not sure it is going to be worth the potential fouls or not.

_____________

My Prediction: (For what it is worth).

Teams of the caliber of 254, 987, 1114 Et al who have access to full field setups at home among a few others I can think of, and have perfected the vision positioning and ball (or tubes for that matter, as far back as 987's 3 tube auto in 2011, as they picked up a tube leaning against our bot without moving our bot even 1/8" in Las Vegas perfectly & repeatedly without fail even once), item /robot seeking/locating already down pat as demonstrated repeatedly in prior years (with plenty of practice time of course), I fully predict now....Will be absolute champs at returning to the neutral zone and plucking or vacuuming, etc. those balls 1 at a time off the rings on the midline without fouling and will do so until there is absolutely no more time left on the clock in the auto period by their very first competition event.

The repeated risk is worth the reward, as it should earn that all coveted Q position #1 spot and Alliance #1 Captain slot.

Those are teams to which the words "You Can't Possibly Do That...It's just too risky" have absolutely no meaning whatsoever in life....That is why they are "Powerhouse and Championship Caliber Teams!"

Now, if one wishes to defend against that type of absolute wonderment, you must be willing to be a really lousy alliance partner(s) (not recommended by this party BTW), and return your robot, after crossing and making your auto shot, back to the neutral zone yourself (or the easiest path back), and block that killer robot's path to the ammunition of more boulders they need to pull off their magic. (Of course that action would be deemed foolish, pretty self destructive, and not very GP).

[cglrcng]

https://www.youtube.com/watch?v=N995FopDHgU

That is beyond simple accuracy....That shot was amazing! WOW! (That wasn't targeted by joystick I don't believe...Was it?)
__________________
Not that many months ago I said personally (while at CMP), that under the 2015 rules, it was impossible to bring in to CMP a bagged robot you built to compete with, tear it completely apart, build an entirely new bot from the ground up using strictly COTS and raw stock carried in at load-in and only the withholding allowance of pre-fabricated parts , get it to pass inspection (I am sure the inspectors kept a good eye out that the rules were followed and fully trust they did so), use 4 freaking harpoons to snatch the 4 cans with, and then appear on Einstein in just 3 days from their starting point or less....But I have a copy of the video proving they could and did absolutely do so. (If only we had the chance on-field to see that masterpiece in action instead of on the sidelines).

Team 2959....That video now sits in a folder right beside that other totally unbelievable harpoon can snatching video...Well done folks!

[hardcopi]

It was by joystick. Our driver (who is a mentor this year) was showing off. It did take him like 3 tries to be fair, but we utilized a camera with crosshairs that year. He was able to repeat the feat a few times that day too. That was after all the off season events too so the robot had a lot of battle scars.

We are looking to repeat that kind of accuracy this year.

[Al Skierkiewicz]

Originally Posted by sciencenuetzel

A question for everyone and maybe it's already understood from previous years. (Year 2 for me)

Is the robots dimensions defined by the frame perimeter? (Meaning in light of this thread if my intake mechanism is 14 in outside of my perimeter is that part considered my ROBOT? Could the mechanism extend over the midline to grab the second ball in a two ball auto? If so would that be considered a violation of the rule as long as my frame perimeter did not cross the midline?

This is a question that can only be answered by the Q&A, however, it is often interpreted that anything that is connected to the robot is the "robot".
That being said, I can think of a variety of pickups that could move the rock without crossing the midline volume.

[Lil' Lavery]

Originally Posted by AlexanderTheOK

I apologize that you found offense in my statement, however, I hope you understand that giving two factually incorrect statements in a row gives off a particularly negative impression on one's memory, so I wouldn't say my statement was particularly out of line.

Neither of my statements was factually incorrect. Please read what I wrote, and not what you think I wrote.

Originally Posted by AlexanderTheOK

I will reiterate my point once again:

In 2013, the margin of error would be defined by the construction and design of your intake, and the teams that were good enough to go for the mid-line Frisbees also had intakes that would successfully capture Frisbees over their entirety. This left such teams with enough clearance that localization via encoders and IMU would yield error that would not affect intake effectiveness.

In addition, the margin of error for getting a foul was that of your entire robot, which would be several feet, meaning if a team WAS to miss the frisbees, they would not accrue a foul. In 2016, crossing the midline in any way is a foul.

Watch the videos of 987 I posted. Their intake was located along the narrow end of their robot, meaning the margin for error was only a couple inches for acquisition. As a result, it meant they had to center their long axis of the robot exactly on the center line as they drove backwards (which is the accuracy you were questioning in your previous post), as they need their intake centered after they rotate. They then had to rotate, and drive a precise distance down the line to acquire the Frisbees, but not overshoot the required positioning to make it back to their starting location without contacting the pyramid (or other robots) in time to fire the newly acquired Frisbees. They executed multiple maneuvers within their autonomous routine that contained the same degree of precision than you're stating is impossible in this years' challenge. That is far from the only example of such precision in autonomous.

[MrForbes]

There is the question, "Can it be done?"

and there is the question, "Should we try to do it?"

The answers might be different for different teams.

[Michael Corsetto]

Originally Posted by MrForbes

There is the question, "Can it be done?"

and there is the question, "Should we try to do it?"

The answers might be different for different teams.

Truth!

[Nathan Streeter]

Coming from a team that usually makes a goal out of capitalizing on auto (2006 shooting auto, 9 auto lines in 2008, 3-5 ball auto in 2010, two-tube auto in 2011, center or corner bridge auto in 2012, two-ball auto in 2014, 3-tote auto in 2015), we're very impressed with the difficulty of coordinating a multi-ball auto this year.

The obvious route; get a ball from the center line would be difficult without the mid-line rule... it becomes astonishingly difficult with it (I'm sure a literal handfull of teams will have one).

The secondary route (achieved four obvious ways) is probably less technically difficult, but requires coordination to get the extra balls from your partner(s)... and only helps if one or both of your partners can't score the ball high in auto (common early on, but some DCMP or CMP alliances may have three High Goal autos).

1. Spybot passes its ball to a HG shooter after it scores its first ball from the courtyard.
2. Spybot receives ball from a partner that passes ball, after the Spybot has shot its first ball.
3. Neutral Zone bot drives into Courtyard, scores, and then comes back for a partner to pass a second ball, then returns to Courtyard and shoots.
4. Neutral Zone bot drives into Courtyard and scores first ball, meanwhile partner(s) pass a second (and third?) ball to scoring bot, who must sequentially score those.

My gut is that these will also be very rare though, since they require partner coordination and may compromise other potential auto modes that would better prepare the alliance to 'hit the ground running' in teleop (such as preparing to defeat other defenses and/or gather the midline balls).

I suspect that an ordinary, reliable HG or LG auto will probably be about as rare (maybe somewhat less so?) as 2-ball autos were in 2014... it's non-trivial to defeat a Defense AND know where you are well enough (or use auto-targeting well enough) to make the shot reliably.

[cglrcng]

https://www.youtube.com/watch?v=r9PvIsZL4cg

Nuff said in that video alone...So can I Al. "Multi-Auto Boulder Game Will Be Confirmed!" (Unless of course the GDC says it isn't so).

If there is a will, and a will to take the risk if paying rewards exist, there is always a way to succeed.

While last years 2015 Recycle Rush rules have nothing to do with 2016 Stronghold Rules or Game...It's Already proven by last years multi-string wire/cable connected units, that anything connected to a robot is considered 1 robot (Unless of course it is someone else's robot that just reached inside your frame perimeter). Technical FOUL! (Batman, Robin & Alfred come immediately to mind...And I'm not talking about the foul part).

Robot defined by the 2016 Stronghold Game Rules:

"ROBOT an electromechanical assembly built by an FIRST Robotics Competition Team to perform specific tasks when competing in FIRST STRONGHOLD. It includes all of the basic systems required to be an active participant in the game: power, communications, control, and movement. The implementation must obviously follow a design approach intended to play FIRST STRONGHOLD (e.g. a box of unassembled parts placed on the FIELD or a ROBOT designed to play a different game would not satisfy this definition)"
_______________
That manipulator is attached to that robot & is a part of that assembly isn't it? Is it performing a task for that ROBOT? If so, then it cannot enter the volume above that midline during Auto according to G13: (Got to just love the unlucky numbering of that rule if 1 tends to be even a wee bit superstitous! LOL.)

Woe is to those poor daring souls who tempt fate & risk it all to perform above and beyond in the Auto mission, and do fail (by entering that forbidden zone, or worse, enters it and touches an opposing robot on the other side at the same time), by points added to the opposing Alliance.

Yes the risks are huuuuge! And the rewards are great. (Huh, WHAAAAAT!?...It isn't a TECH Foul though?? Just a simple Foul, and a further robot touch an additional simple Foul? you suffer 2 penalties=Net zero if you make the shot? And you don't even lose the resulting shot if made.....WOW!

Do the math. Auto Boulder worth 10 Points Each in Top Goal....Foul adds how many points to the opponents score? You say 5 points =Net gain, loss or tie? Risk vs. Rewards....Boulders on Midline value in Auto 2016 = Original Litter value in 2015?). I'LL TAKE ALL 6 PLEASE...One perfectly placed boulder at a time. Math is SOO GOOOOOOD.

I can already hear some begging, "errr lobbying" heavily next week, for the Auto Period to be increased with 6 available boulders out there on the oh, so dangerous MIDLINE. The math and the risk takers thinking outside the box just demands it!

Hint: Now, if you know someone is going for a multi-boulder auto, and you are on the opposing alliance and are NOT, a better strategy would be to simply add a delay of time to your auto mode programming of any possible movements toward crossing and shooting or delivering until a wee bit later (rather than earlier in the auto period). Nothing anti-GP about making the risks greater for the opponent by doing absolutely nothing for a wee bit longer. Just to watch em' sweat, of course still attempting and making yours, just not as early in the period.

3.4.4 AUTO Rules

G13 During AUTO, ROBOTS may not enter the volume above the MIDLINE. Violation: FOUL. If contact is made with an opponent ROBOT beyond the MIDLINE, an additional FOUL is assessed and the opponent ROBOT is immediately awarded the CROSSING of the closest DEFENSE from the point of contact.

MIDLINE DEFINED:

MIDLINE a line of 2 in. white gaffers tape that bisects the length of the FIELD
_______________________


The hunt to copy and quote of which leads to another question:

Saw the following (asking the question based on the exact terminology used in the rule, not attempting to lawyer it, just was curious on the exact wording and curious as to the answer that would be given by the GDC if asked). For just discussion, what say the parties here on CD?

G26 At any given time, only one (1) ROBOT may contact the carpet or BATTER inside their COURTYARD. Violation: FOUL. For every five (5) seconds in which the situation is not corrected, TECH FOUL

The question: (Given that they are talking about say Blue Alliance entering their own COURTYARD to play defense and defend their Tower and Castle against the dastardly attacking RED Alliance, and not the other Courtyard of The Dastardly Red Alliance, The Blue Alliance are actually attacking...Or, vice versa).

Does that actually mean that 1 Blue Robot could be on the BATTER of The Blue Tower, and 1 Blue Robot could be on the carpet of the Blue Courtyard at the same exact time, as long as they do not actually both occupy (contact), either The Batter or The Carpet at the same exact time? (Otherwise, they could have just written the rule as simply "Only 1 Robot can enter or occupy their own Courtyard at a time to defend or play defense against the attacking alliance" or the addition of 1 word "either" between the words "contact" & "the"). Since you cannot get to the Batter without entering the Courtyard.

Just more food for thought.

It would be quite easy to play tag team (one parked on the batter in front of a low goal & one on the carpet running circles around an attacking bot in active defense). You couldn't tag team Batter to Carpet, Carpet to Batter (as that would require both bots at some point (contacting), batter or carpet at the same time...But, you could Courtyard to Secret Passage, Batter to Courtyard Carpet, and back to Batter/Courtyard. Either way 2 bots would still be in their own Courtyard at the same time playing defense against an attacking alliance.

Only the GDC knows the answer. But, someone actually playing the game would have to ask the question(s). (The use of the word "or" in both letter & spirit of the rule may just mean simply the same as "either." Or, it may not.

[cglrcng]

Originally Posted by MrForbes

There is the question, "Can it be done?"

and there is the question, "Should we try to do it?"

The answers might be different for different teams.

Oh so true.

But for those capable...The math simply demands it. Even with 2 penalties suffered, as long as you make it High Tower Auto Goal there is a net zero for the attempt. 1 penalty suffered you still gain net 5 points, no penalties 10 points, each attempt.

_______________

Added: ".......The answers might be different for different teams.[/quote]"

Well, when it comes to Harpoon Type Can snatchers...It certainly depends on who is standing just a couple of feet away on the sidelines!

[Dillan_McDonald]

It is viable, you would have to start in front of the low bar or be able to precisely scale/cross another defense, drive forwards, shoot, come back, pick up the ball on the middle line in front of said obstacle, drive back to the shooting position, and shoot again. It is all possible if you have a fast robot with very precise and accurate movement and shooting capabilities, however it would be very close on time.

[SoccerTaco]

Originally Posted by AlexanderTheOK

While I would hope Q&A would change this rule to lower the potential risk of going after a second boulder, the rules as they are give a margin of error that is nearly an order of magnitude lower than that of previous years.

Hail Fellow. I greet thee!

While I definitely do not expect to see many teams do a 2 boulder auto - it will be extremely hard in 15 seconds - I also don't think it is quite as impossible as you make it out to be to get a boulder off the midline without obtaining a foul. Consider this scenario which I do believe is all perfectly legal.

Robot is placed on the field at the back edge of the auto zone, ~12 inches from the midline, preloaded with 1 boulder.

Auto starts.

Robot sets their boulder down in the auto zone, so they no longer control any boulders.

Robot lowers a roller arm intake mechanism toward midline. The team strategically placed their robot on the field so that the intake will contact the midline boulder without crossing the midline.

In this scenario, the robot has not driven one inch, and they have a boulder off the midline. The rest is obvious - cross defense, shoot, return to auto, collect the boulder that was set down, cross defense a second time, shoot.

I'm not saying this is easy!! But that is mostly because of the time limit, not because of the midline foul risk.

Godspeed on your quest.

- Steve

[BrianAtlanta]

I went through the weekend thinking about 2 ball auto, and not seeing a way to do it. After seeing this thread, if you accept the 5 point foul, it would be possible for a 2 ball auto. It comes down to cycle time, move faster or move less. I can imagine some designs that would work, but it's not an easy undertaking.

We'll see what our students come up with and they want to do. If for nothing else, it's a nice exercise to see what's possible.

Thanks for inspiration to challenge my view of what possible.

Brian

[cglrcng]

Originally Posted by Dillan_McDonald

It is viable, you would have to start in front of the low bar or be able to precisely scale/cross another defense, drive forwards, shoot, come back, pick up the ball on the middle line in front of said obstacle, drive back to the shooting position, and shoot again. It is all possible if you have a fast robot with very precise and accurate movement and shooting capabilities, however it would be very close on time.

______________________________
Or, start in the Spybot position, shoot/score, crossover an easy Defense (Sally Port and Drawbridge would be easiest to exit the Courtyard (if avail.), unless your bot is less than 16" high which is the height of that Low Bar Defense). Get and load Boulder go back across shoot & score (rinse and repeat until 15 seconds are up).

**And if the Drawing at 1.2 Introduction (labeled Figure 1-1: FIRST STRONGHOLD Arena), is correct as I suspect it is, after seeing the vid's and other official docs.

And, if you read G20 in the literal sense. Then, you cannot go from the Courtyard you are attacking through the Secret Passageway & directly back to the Neutral Zone at any time. That is "your opponents secret passage." Unlike the 2-2 Figure appearance, wherein it looks like that is your secret passage if you look at the flag colors.

This would make you only go back and forth (while attacking at least), between the Courtyard and Neutral Zone via crossing the Defense structures of the opponents Outer Works you are attacking and attempting to Breach.

This does make it a whole lot harder to do the repeated multi-boulder auto attempt (instead of going through what appeared to be by all of the direct overhead field illustrations like Figure 2-2: Zones & Markings, Figure 2-4: OUTER WORKS DEFENSE Placement, among others (that appear to have the flags reversed in all of the direct overhead views), and made it easily appear that you could legally (w/ the G50 Rule), just make a clockwise legal circular driving motion of Neutral Zone, over the opponents Outer Works/Defenses into the opponents Courtyard, attack the opponents Tower, exit the opponents Courtyard to YOUR SECRET PASSAGE -just going twice over that much lower only 1" high X 3" wide metal strip, then back into the neutral zone, snatching 1 ball at a time by plucking it off the line, and keep making that circle repeatedly pretty much without interference, as long as your alliance partners knew what you were attempting to do in auto, and stayed out of your way after their attempts.

3.4.6 ROBOT to ROBOT Interaction

G20 ROBOTS may only enter or exit their opponent’s SECRET PASSAGE from/to the opponent’s COURTYARD. A ROBOT is considered to be within the SECRET PASSAGE once the only carpet the ROBOT is in contact with is the carpet inside the SECRET PASSAGE. Violation: FOUL. If repeated, TECH FOUL

Emphasis mine.
****This makes the entire game that much more difficult to say the very least. Most people will easily parse the game and strategize on an overhead field view like Figure 2-2 more so, than to do so, on an angled far off view like Figure 1-1.

I could not figure out why I was getting a headache on the first read through the rules on Saturday morning (when I was switching back and forth between vids and the rules in PDF format on my laptop (later on I was reading them in a BW printed copy instead of color, and I thought it was just the flipped opposite field ends and that is why it wasn't making any sense (I just could not visualize what was making me so uncomfortable and I just couldn't understand why, thought I was just tired (the action actually gave me a headache trying to make sense of that field and the confusing aspects).

No, it was that the direct overhead drawings were all screwy due to the reversed flag colors, that made it appear the secret passages were Red w/ Blue Castle & Blue w/Red Castle instead of the much more proper Blue one side of field/Red on the other and only the attacking robots flipped to the opposite side of the field as attackers, and the much more natural human players w/ drive teams, and Spies on the far side, away from their natural and associated drive teams actually spying.

I'll still say the attempts will still be there from the usual teams, but it won't be as easy as I thought it would be earlier without that easier faster clockwise driving allowance.

It now appears like exiting from opponents Secret Passage to The Neutral Zone earns you a Foul and if you repeat the action it would be a TECH Foul also. (While G20 is 2 complete sentences, it is one section ending with the violation and penalties for violating, but, specific to G20 only, you cannot violate the 2nd sentence as it is descriptive only & not actionable to you, you can only violate the first sentence - Highly descriptive/very limiting/easily violated/& highly actionable). Anyone else read it that way too?

****Yes, I also realize there are other rules that also follow concerning the Secret Passage and Robot Interaction that follow....But, those have different meanings/violations/penalties.

[AlexanderTheOK]

Originally Posted by BrianAtlanta

I went through the weekend thinking about 2 ball auto, and not seeing a way to do it. After seeing this thread, if you accept the 5 point foul, it would be possible for a 2 ball auto. It comes down to cycle time, move faster or move less. I can imagine some designs that would work, but it's not an easy undertaking.

We'll see what our students come up with and they want to do. If for nothing else, it's a nice exercise to see what's possible.

Thanks for inspiration to challenge my view of what possible.

Brian

This is the only way I see teams doing this consistently, just taking the foul and going for the 27 pt autonomous.

Although some users have already shown that there are in fact some limited markers on the field to localize on (the auto line specifically would be my go-to when I see if my team wants to attempt this) and remove accumulated error. I could see a team or two using these and getting right up against the midline now.

0-32 in 15 seconds doesnt have quite the same ring to it though