[MCC] Minimum Competitive Concept 2013

[IKE]

Each year I am amazed by what teams come up with to compete in FRC. Teams have a ton of wonderful ideas and some even see good execution of those ideas.
I would like this thread to focus on the "Minimum Competitive Concept" for a robot for 2013. It is often easy to identify all the possible tasks you could have a robot do. Prioritizing those tasks, and realizing it in the form of a competitive robot is in my opinion much more impressive.

If you haven't read the Simbotics Strategy Presentation (http://www.simbotics.org/files/pdf/effective_first_strategies.pdf), please do before responding to this thread. Especially review the "Golden Rules 1&2".

Assumptions are that one of the priamry goals of the MCC is to play in elims (not necessarily win on Einstein), and you team has mid-pack to lower fabrication resources.
Please list your assumptions, strategy to seed high, estimate of a winning score, and what robot design elements would achieve this score.

I will throw my $0.02 in later.

[avanboekel]

This is always easier to do in hindsight, but here goes my MCC for the 2013 season...

Drivetrain-Simple tank drive, 4wd or 6wd depending on if the team were to go long or wide.

Shooter-Fixed angle straight shooter. This type of shooter is easier to machine, and can be made in a matter of hours. To change height of shot, the team simply changes rpm of wheel. The shooter needs to be low so it can be fed from the bottom slot.

Hanging. Passive climber much like 2338 had (we articulated ours up and down so we could go under the pyramid, but the idea is much the same). No motors required. Guaranteed hang every time.

[Boe]

Heres what i think


6wheel kit bot with 4 CIMS
60 inch robust blocker that deploys to 84"
Simple and quick ten point hang
smart drivers that dont commit fouls
2/3 disc dump into 1 pt goal in auton

[Andrew Schreiber]

MCC 2013 - Fixed angle shooter (bucket loader). Drops 2 into high in auton. (50% accuracy). Runs 3-4 cycles of human load. Pneumatic 10 point hang.

I would assert this robot, if driven properly, would be in contention at most regionals. Furthermore, I would also assert that the KoP drivetrain, FRC3313, and Ri3D gave every team the info needed to build this robot.

Even at 50% accuracy: 34 points isn't bad (assuming hang)

[Pault]

This year seems very easy to figure out the mcc.

1. Reliable drivetrain
2. Effective feeder station loader
3. Fixed angle shooter
4. Consistent 18 point autonomous
5. 10 point hanger

I am torn over whether limboing should be in this list. It is not necessary to be be competitive if you are good at everything else you do, and makes hanging harder, but it is fairly easy to accomplish and really makes you better at defense, as well as beating defense. So, I am going to have to say yea for now

6. Limbo

[Nemo]

Kit drive + 3 disc autonomous shooter in high goal + passive 60" blocker + passive 10 point hang. No teleop frisbee loading.

If it worked properly and was driven well, I'm pretty sure it would get into the elims more often than not. This is a solid 2nd round pick.

Given that Ri3D showed everybody one way to make a solid shooter so early on, I think it's fair to include a 3 disc auto in MCC.

[Chris is me]

This is easy. The Robot in Three Days, or 3313.

This year's "minimum" is way higher than previous years due to the relatively easy game task. I would say the "minimum" robot would be a fixed, simple human loaded bucket shooter (linear, curved, does it even matter?) that could auton, cycle three times, and hang (passively is probably best for MCC). 2 point or 3 point goal isn't that important but I don't see many teams having trouble over six weeks tuning their robot to be accurate enough to use the 3 point goal.

This concept (the "pure cycler" distilled into its simplest form) is potentially competitive at all levels of play, falling off only at the Championship level, but then you can add a net and still be in the hunt.

[JesseK]

My additional limitations for the MCC
- No Pneumatics
- Very simple code-wise
- Limited budget

Basic: Plywood circle-shooter with a ram-style hang. Feeds from chute over the shooter into a plastic-bucket-hopper, into a 90-ish-degree 8" treaded shooter. Shoots from either rear corner of the pyramid.


KOP Drivetrain that uses 2 CIMs total. Add more depending on available budget.
Shooter: Plywood wheel with the 3M friction tape purchased with PDV from AM, directly-driven by a CIM. Shooter channel uses 1/4" thick x 1" wide strips of squishy silicone rubber on the outside to provide friction for spin and plenty of cushion while the disc is under compression. Shooter is adjustable while in the pits (thinking 3/8" all-thread) so the exact angle can be tweaked, but is static while the robot is live.
Hopper: 5-gallon bucket, one with a very rigid rim support trellis on the outside (useful for mounting)
Indexer: Any number of motor/gearbox combinations would work here. Banebots RS-775 + 10:1 Versaplanetary would be my choice in order to spread the PDV's around. The dual spiral arm would be made out of wood and would kick the bottom-most disc in the hopper into the wheel. Code would tweak the % power sent.
Feeder chute: Tough call for what material -- need it to be smooth (various woods would work) but also be able to absorb some amount of energy from a falling disc (polycarb with some flex). Access to a heat gun & patience would make polycarb win. Just make sure the bottom isn't concave. Mounted on top of the shooter, it has a fixed angle with the correct height for the middle feeder chute. Is doable in such a way that the robot can still go under the pyramid (useful for avoiding defense).
Hanging: ram-style hanger. I didn't do this one during the season, but saw plenty of robots do it right and plenty do it wrong.
Bumpers: (yea, that's right!) Letters are all-white and are spaced according to various calibration points for lining up shots. That way the driver knows the range of depth where a shot will hit vs miss. Simply put the pyramid corner on one of the numbers.


4-5 cycles, 18-point autonomous, 10-point hang. With practice, could lead an alliance into elims. Without practice, the team would easily make a great off-side cycler.

----
IMO, the problem with using the de-facto defense box-on-wheels-with-pool-noodles robot is there are plenty of alliances with FCS's that would not what that type of robot unless the robot was GREAT at pushing others out of the way. That means 148lbs, (probably) 2-speeds, and an extension for the FCS blockers such that they can be out of the way. This adds pneumatics, which an MCC team (IMO) probably wouldn't understand well enough to maintain over the course of a season.

[avanboekel]

KOP Drivetrain that uses 2 CIMs total. Add more depending on available budget.

IMO, a 4 CIM minimum drivetrain is a must, especially for a MCC. When my team is looking for a second round pick, this $60 choice would be the difference of them making the eliminations, and packing up early.

[Mike Marandola]

6 wheel kit drive train, fixed pneumatic wheel shooter, ability to load from the hp, passive 10 point climb, great drivers.

[Nemo]

A robot with 3 disc auto (67% accuracy = 12 pts), 3 cycles (50% accuracy = 18 pts), and 10 pt hang (80% success = 8 pts) contributes 38 points on average.

That 38 point robot is one of the 8 best robots at most events, and it's in the top 16 in nearly all events. The robot doesn't need to be that good to sneak into elimination rounds most of the time.

[EricLeifermann]

see 2169

[Adam Freeman]

This is an interesting year for an MCC, especially with how the game played out. In this case and MCC could actually be a very very competitive machine at the majority (if not all) levels of competition.

It's almost an 1114 (minus climber) / 862 hybrid (lite) machine.

- KOP drive - Wide configuration.
- Single Speed 4 or 6 wheel drive. 4 CIM Toughbox/Mini-Toughbox transmissions.
- Passive 10pt hanger.
- Human player loaded - Front loaded (similar to 1114).
- Roller fed, single wheel 1/4 arc shooter powered by a CIM/Mini-CIM.
- Fixed angle.
- 2pt, maybe 3pt goal capability.
- No speed control, just % based shooter control.

Almost any team could build this machine....and it would be incredibly effective as a secondary scorer or, for sure, a third machine. It seems odd that an MCC could actually be this simple, yet this effective.

And with a couple of more advanced tricks, such as a non-passive hanger to be able to get under the tower and a higher level drivetrain (shifter or 6 CIM single speed) and you have a World Championship level machine with a great driver (*610 anyone?).

*As a side note, I have not actually seen 610's machine, but I imagine it's simplicity is what made it very effective.

I kinda wish I had this perpective at the start of the year. Would have led to so much less burn-out.

-Adam

[Lil' Lavery]

Building off of what Nemo said, a lot of your MCC concepts seem much more like early 1st round selections and alliance captains. 18 point autonomous modes? Three, four, FIVE(!?!) cycles?

Until week 5, the mean elimination score for each week was beneath 100 (with it basically sitting around 80 points until week 4). Week 5 and 6 was right around 100. The median elimination score for the year was substantially beneath 100. The goal isn't to win a regional, but simply play in the eliminations.

A robot that can contribute in the low-mid 20s offensively (ie, autonomous and 10pt hang) and a bit of defense is more than adequate to reach the eliminations at a majority of events.

[AdamHeard]

Building off of what Nemo said, a lot of your MCC concepts seem much more like early 1st round selections and alliance captains. 18 point autonomous modes? Three, four, FIVE(!?!) cycles?

Until week 5, the mean elimination score for each week was beneath 100 (with it basically sitting around 80 points until week 4). Week 5 and 6 was right around 100. The median elimination score for the year was substantially beneath 100. The goal isn't to win a regional, but simply play in the eliminations.

A robot that can contribute in the low-mid 20s offensively (ie, autonomous and 10pt hang) and a bit of defense is more than adequate to reach the eliminations at a majority of events.

Agreed.

[lynca]

Assumptions are that one of the primary goals of the MCC is to play in elims (not necessarily win on Einstein), and you team has mid-pack to lower fabrication resources.

MCC is a great road-map for many 2nd tier teams. We are definitely a 2nd tier team in Texas and found success with a MCC robot.


Please list your assumptions, strategy to seed high, estimate of a winning score, and what robot design elements would achieve this score.

Assumptions:
zero machine shop time (hand tools only) , no welding
occasional laser cutting (varies on donations and friends )
partial field to practice (no pyramid)

Strategy
shoot 3 point shot from one place (use physical hardstop to align)
passive climb (no actuation , can't build a pyramid , so climbing is out)
passive feeding of frisbees
be tall enough to not be blocked

Design
6-wheel VEXpro single speed drivetrain
tall fixed angle shooter (KISS)
passive 10-point climbing mechanism
gravity fed hopper

Accomplishments
1. LSR - Regional Finalist , Engineering Excellence
2. Alamo - 2nd Ranked team, semi-finalist
3. Ranked 5th in Texas ( unofficial rankings (http://www.chiefdelphi.com/forums/showthread.php?t=114974) )

Full Detail and videos of our robot. http://2013.discobots.org/node/72

[EricLeifermann]

Building off of what Nemo said, a lot of your MCC concepts seem much more like early 1st round selections and alliance captains. 18 point autonomous modes? Three, four, FIVE(!?!) cycles?

Until week 5, the mean elimination score for each week was beneath 100 (with it basically sitting around 80 points until week 4). Week 5 and 6 was right around 100. The median elimination score for the year was substantially beneath 100. The goal isn't to win a regional, but simply play in the eliminations.

A robot that can contribute in the low-mid 20s offensively (ie, autonomous and 10pt hang) and a bit of defense is more than adequate to reach the eliminations at a majority of events.

Though true, shooting this year was really easy. 2169 doesn't get much simpler. Yes they were super competitive and Galileo finalists. But their robot was a 1 wheeled curved shooter and a 10 point hang. They made 0-2(if blocked, though very rare) runs per match and scored 3 discs in auto.

The MCC, almost any form discussed here, this year could have won Einstein.

[CalTran]

*cough*610*cough* Minus the Coyotes' tray extension and their variable shooter, they seem pretty darned simple for this year. The Logomotion 1503 of this year, if I may.

[efoote868]

I think Robot in 3 Days showed us how to do this pretty well this year.

I also think that if you removed the 10 point hang, you'd save some complexity. It would also make the team's decision whether to go for another 12 points in discs or a 10 point hang much easier.

[Siri]

Building off of what Nemo said, a lot of your MCC concepts seem much more like early 1st round selections and alliance captains. 18 point autonomous modes? Three, four, FIVE(!?!) cycles?

Until week 5, the mean elimination score for each week was beneath 100 (with it basically sitting around 80 points until week 4). Week 5 and 6 was right around 100. The median elimination score for the year was substantially beneath 100. The goal isn't to win a regional, but simply play in the eliminations.

A robot that can contribute in the low-mid 20s offensively (ie, autonomous and 10pt hang) and a bit of defense is more than adequate to reach the eliminations at a majority of events.I think this comes from a mild discrepancy in what MCC is intended to mean. Many people are taking it as 'what shouldn't be that hard', while others are following the OP's 'how to play Saturday afternoon'. I think it's uncommon for the former to end up so much less than the latter (very cool game!), so it hasn't come up much in the past.


I'd say to play Saturday afternoon, the bare minimum you can do is put the kitbot on steroids together, put a solid but light (low CG) blocker+passive hang on it, and buy an extra set of wheels. Then proceed to drive the original wheels off for the remaining 5 weeks. Know the rules really, really well. You'll hear your name before lunch Saturday at nearly every qualifying event. (Don't have a practice field? Find a parking lot--that's what the extra $40 in wheels is for.)

The next step up from this is to mount a low autonomous-only shooter in front of the blocker+hanger. Do the math: don't move to 3 discs or the high goal if it lowers accuracy too far. Work on human feedings if there's time, but don't sacrifice drive time for it (if the goal is to play Saturday afternoon).

[Alpha Beta]

Making eliminations required a higher level of proficiency this year than most. We were in a position to make the #24 pick 3 times this year and twice we were amazed to find a robot still available with an autonomous, great drivetrain for defense, weight available to add a full court blocker, and a reliable 10 pt hang. When making our pick list for our last picks the following were our desired characteristcs in order:

#1. drivetrain and driver skill
#2. autonomous consistency from a non-center position
#3. weight and mounting structure available to add a full court blocker if not already built into the robot.
#4. fast hang/climb (slow climbers spend too long not playing defense)

Hang was further down the list because it required the robot to break off a devensive assignment, where a solid autonomous did not.

Now to make eliminations at champs was a whole different ball game. We had a bevy of quality full court shooters with auto and reliable hangs to choose from with the 24th pick. 2471, 180, and 503 were all interesting options in our scouting list before the Holy Cows made their final choice.

TLDR: The quality of robots this year raised the stakes considerably on what did and didnot make the final 24 for eliminations.

[IKE]

So My MCC before the comeptition season started was quite a bit different than what I am seeing here.

First of all, I usually try to set my MCC up so that it will score at 2x-3x the median team. My estimates for the median team before competition season started was 10 pts. I do this by thinking through what I expect the top OPR for the year to be (my estiamte was right around 100 for this year and ended up pretty spot on). I have then noticed that the median team OPR tends to be around 1/10th the top OPR. or about 10 points (median week 1 this year was closer to 13 pts, so I was off quite a bit). this lead to my 2x-3x robot needing to score about 20-30 points. The reason for the 2x-3x is that if you want to seed high enough to be a picker, you need to win most of your matches. At 3x the median, you should win at least 75% of your matches as your robot alone will be equal to the average opponent score you face. My 30 point estiamtion of this score proved to be a bit off.

So, what was my thoguht on how to most reliably get to 20-30 points... The 1 point goal and 10 point hanging. My thought was to build a 1 pt. dumper that could score:
2discs in auton driving forward, hitting the goal and dumping in 2 discs for 4 pts. Next, this robot would make 4 trips ferryign its 16 discs into the 1 pt. goal This should be pretty easy to make those 4 trips as it is relatively easy to line up no one would be defending a 1 pt. dumper... 16 pts.
Then execute a 10 point hang.
This robot strategy only uses the teams share of discs (15 whites plus 1 colored disc) and would be a very solid 20-30 pt. (4+3*4*1+10=26, 4+4*1*4+10=30).
Lastly, it is a pretty simple bot, so adding a blocker is no big deal.
*************************************************
So looking through the data from 1114 database, the median when looking at teams BEST OPR for the season was 13.78. The median OPR for Auton when looking at best was 4.01.

The robot stratgey assigned above would be fairly comeptive, but would not meet my 3X median desire. I also would not have growth capability to get there. This was mostly due to the median being higher than I was expecting. Looking over legacy data, I will probably use a 7X instead of 10x delta for my best vs. mdeian.

The 26-30 pt. window when compared to OPR is 75-80 percentile. This is actually quite impressive when you think about it.

Of course this also requires a high level of execution of that particular design. I know of 1 team that executed such a design, and their OPR was 11.21. This was largely due to not having an auto (which is -4 points), missing a good chunks of their hangs (their climb OPRs were 2.9 1st event and 7.9 2nd event). This team was also frequently asked to play defense as they were not deemed a strong scorer. Because of this, they typically only did 1-2 vollies (4-8 pts.)

They were however 9th and 12th picks at the two events they attended, and were ranked 12th and 29th at those same two events. Had they gotten to where they were doing 26-30 points, they likely would have been in the top 8-10 scorers at both of their events. Not too bad for a 1 pt. dumper with a 10 point climb.

This concept would not have done well at an event like MSC. Because it was fully limited to 30 points, it would have topped out early in the season.

[Clinton Bolinger]

MCC = 862

- Passive 10 point Climber
- Passive pyramid alignment for shots(backside of their climber)
- Fixed Angel Shooter
- Human loaded from Feeder Station

Added for Champs
- Passive 60" Blocker
- Simple pneumatic 84" block that could be deployed at will

2 District Wins, 2 Engineering Inspirations, & Division Win for a team that hasn't won a blue banner in the history of their team.

I think this season 862 let Strategy Dictate their Design.

-Clinton-

[Chris is me]

The problem I have with the MCC being a one point dumper is twofold. First, it's not "competitive" at any level of higher play. Reducing the disc count by 16 just to score 16 points just won't cut it at the deepest events. I would rather teams shoot those 16 discs at 50+% accuracy into the two point goal, honestly. Secondly, I saw a lot of solid 1 point dumpers with hangs that did 3 cycles at week 1 / 2 events. Maybe two of them made elims. If you can't make elims that early in the season, there aren't a lot of ways you'll make elims later into the season.

I really think the "minimum" starts with the 2 point goal this year. Make 8 in auto, 6 points a cycle, and hang. Then again, I define "competitive" as "at least has an outside shot of making elims at any event"... maybe that's too high.

[IKE]

...
I really think the "minimum" starts with the 2 point goal this year. Make 8 in auto, 6 points a cycle, and hang. Then again, I define "competitive" as "at least has an outside shot of making elims at any event"... maybe that's too high.

I don't disagree with your statements. I did think that 30 points would have been more competitive this year than it ended up being. I felt I needed to mention what I thought would be a minimum competitiv concept before comeptition season came out. All that being said, I did not see any 1 point dumpers that delivered at the level I mentioned. It may be a 1 pt. dumper, but it would be a very good 1 pt. dumper.
********************************************
What point level does an MCC need to be at this year? I generally say that the top 16 scorers for an event are pretty solid picks for playign in Elims. This would have been a going in high OPR of 56 points for Archimedes... It was 51 points for Newton.

[Nemo]

So, what was my thoguht on how to most reliably get to 20-30 points... The 1 point goal and 10 point hanging. My thought was to build a 1 pt. dumper that could score:
2discs in auton driving forward, hitting the goal and dumping in 2 discs for 4 pts. Next, this robot would make 4 trips ferryign its 16 discs into the 1 pt. goal This should be pretty easy to make those 4 trips as it is relatively easy to line up no one would be defending a 1 pt. dumper... 16 pts.
Then execute a 10 point hang.
This robot strategy only uses the teams share of discs (15 whites plus 1 colored disc) and would be a very solid 20-30 pt. (4+3*4*1+10=26, 4+4*1*4+10=30).
Lastly, it is a pretty simple bot, so adding a blocker is no big deal.


Team 2531 had this strategy at Minnesota North Star. They were good for about 25 points per match, and they ended up as alliance captains. They were pretty solid.

[KrazyCarl92]

It's most useful to think of this in terms of what the top teams were looking to select for partners, since ultimately it was their decision as to who made elims.

My MCC robot based on this idea:
-All wheels driven in drivetrain, 4 CIMs, and preference to robots with all traction wheels. Let's be honest...in this day and age when this describes the kit bot drivetrain with 2 extra CIMs (in gearboxes set up to accept them no less), the fact that there were plenty of teams with undriven wheels or casters is disappointing. Crossed many a team off of our lists for this reason this year.
-Score somewhat effectively, somehow, in auto. Shoot 3 in the high goal, 2 in the high goal, 3 in the middle goal, 2 in the middle goal, or even 3 in the low goal. I don't care as long as there are points on the board. If it's more on average than the next best option, your number gets called.
-Active 10 pt hang. I personally think a hook on the end of a pneumatic cylinder is simpler to construct than a passive climber based on prototyping from this season. Plus, some teams struggled with the first rung of the pyramid being lower than anticipated, so that negatively affects passive climbers while only affecting those active climbers who were cutting it close as is.
-Ability to add a full court shot blocker to go to >80". Preference given to teams that already have a FCS blocker...I trust a demonstrated robot trait above my ability to ad hoc something any day.

That's it. A 16-28 pt per match robot depending on how it scores in auto and what kind of consistency it can achieve. No human loading, since if anything this robot would take disc scoring opportunities away from better teleop disc scoring partners. The robot could just play D for the entirety of teleop until it decides to go hang.

[cadandcookies]

Team 2531 had this strategy at Minnesota North Star. They were good for about 25 points per match, and they ended up as alliance captains. They were pretty solid.

[insert disparaging comment about the depth of Minnesota regionals here]

[Dragonking]

4 cim, 4-6wheel tank drive, low cg, short (under 30in) or 60in blocker
optional 84in blocker
passive feeder slot collector
90 degree or linear shooter

2/3 auto in either 2 or 3 pt goal (8-12pts)
2 cycles of 4 discs in 2 or 3 pt goal at 75%(12-18pts)
10 pt active or passive climb (10pts)

Total 30-40pts

Should be competent defenders and smart drivers.
Decent Scouting team.
Wins at least 1 award.

Should be low seeded captain or first round pick at most regionals. 2nd pick at deep regionals. Guaranteed elims not counting worlds, Michigan, or IRI