Finger Lakes Regionals: FIRST Robotics 2 Competition Info

[ideasrule]

Since there are quite a few robotics people here, I'm wondering if anybody is at the Finger Lakes Regionals. If not, when's your next competition?

 

[Integral]

Team 957 placed 33rd in a field of 61 at the Portland Or Regional.

Here is our high point,

 

[Chi Meson]

Blue Alliance videos are not available yet. Tell me tell me tell me I have got to know...

How many teams bothered with hanging at the end? Did it prove to be beneficial or a losing scheme re points? Did any teams hang from other robots?

And Integral, is there a story there in that picture? Why is your bot all alone?

 

[ideasrule]

How many teams bothered with hanging at the end? Did it prove to be beneficial or a losing scheme re points? Did any teams hang from other robots?

The competition really shattered my expectations. Scores were very low, around 4 or so for each alliance, and matches with no penalties were extremely rare. Scores of 0 were common in the early seeding matches as teams tempered with their robots. Not many teams managed to score 2 goals in the finale. Contrary to what I expected, a successful hanging mechanism could easily decide the outcome of a match.

Also contrary to what I expected, a successful autonomous was enormously helpful. Getting balls into the right zone was both crucial and hard to do during teleop because even in the opening seconds of teleop, teams usually had more urgent tasks to carry out.

As for the teams themselves, quite a few (25%?) had some hanging mechanism, but only a handful had a reliable mechanism that actually worked. There was one robot that could extend a ramp for other robots to get on the platform, but that robot wasn't good at anything else, and I didn't see any robot successfully take advantage of the ramp. No robot could suspend itself from another robot. Only about 30% of the teams had a working autonomous. Our team could autonomously score from the offensive position, but one team we played against in the finals, team 217, once scored from the defensive position and nearly scored the other 2 balls. The only reason they didn't was because they were too close, so all the robot needed to do was drive backwards a bit before shooting and they would have gotten 3 points in autonomous. That was a frightening prospect, but for some reason 217 didn't score anything during autonomous during the finals match.

Other observations: A ball possession mechanism is crucial, even more so than a good kicker. An adjustable kicker, even if it has only 2 settings, is very useful. A common penalty was the "ball penetration" penalty because balls kept on getting below robots, so some way to avoid that is useful.

 

[Integral]

CM,
So you have yet to compete?

Few teams had arms and fewer made any effort to hang. Virtually none of the bots in the finals made any effort to hang or even had arms for hanging. I think our drive team wanted to hang at the expense of regular game play, it was a detriment to our team. The only match we did well in was the match in which we hung as well as scoring our only point. We have a very nice kicker which in my opinion was under utilized.

I may have mentioned that the team I mentored for last year merged with the team from the cross town rival HS. Their head mentor was a Woody Flowers award winner 3 years ago was reputed to be good. I have been less then impressed and am now determined to remove the reigns from his hands.

The weekend held many low points for me but the worst was our last match, when following the hanging match mentioned above our head mentor was convinced to take the loosing side in a 6v0 game strategy, attempting to run up the score differential to game the ranking system.

This decision stunned the part of our team sitting in the stands. These kids were wanting to see the bot perform well and actually be on the winning alliance in a match. Something neither team did last season.

The match started great, we were playing zone 2 and opened with a beautiful kick which was a narrow miss in autonomous mode. Then beyond all belief our bot turned around and started booting balls at the OPPOSING goal. At least one of our students was driven to tears by this move.

Our illustrous head mentor.

 

[Chi Meson]

Head DEmentor looks like.
(heh heh)
(Don't tell him)

His plan was absolutely so very NOT "GP," and you know what I mean.

 

[Chi Meson]

CM,
So you have yet to compete?

We are only competing in the Northeast Utilities Connecticut Regional this year (and then three or four post-season events) which is week 5. So it is a long wait. We wanted to also compete in the WPI event (Woostah), but we didn't have enough money in time for it. WPI would be next week. Around here, the first weekend event is New Hampshire, which is much sought after (Kamen's home-field). We are partially sponsored by Northeast Utilities, so we "must" attend the CT regional if that is to be our only event.

As I said, a long wait ahead.

We have a hanger which should be successful. On the last weekend, we found out that the worm-drive motors were either too slow or too weak to lift our bot, so we quickly changed it over to a CIM and then installed a braking mechanism. We never got to check it out, since we were still fitting it into place on the shipping date.

Unfortunately, we decided to "withold" the kicker, thinking it would be the most crucial piece to work on during the interim. Up until this moment, the rest of the team was thinking that goal-scoring would be much higher.

 

[ideasrule]

His plan was absolutely so very NOT "GP," and you know what I mean.

I disagree. The whole point of the scoring system is to encourage teams to use novel strategies. On Chief Delphi, a lot of teams were talking about the best strategy to use. The mentor's strategy--to score for the opposing alliance--was one of the most frequently mentioned ideas. Another was to agree with opponents that neither side would defend excessively, so that both teams would score a lot of goals.

 

[Chi Meson]

I disagree. The whole point of the scoring system is to encourage teams to use novel strategies. On Chief Delphi, a lot of teams were talking about the best strategy to use. The mentor's strategy--to score for the opposing alliance--was one of the most frequently mentioned ideas. Another was to agree with opponents that neither side would defend excessively, so that both teams would score a lot of goals.

I'll accept that as an argument, but I really believe that the "shoot-the-moon" strategy is counter to the game's intent and philosophy. I personally would never endorse it.

 

[Integral]

Am I misunderstanding something?

It seems the best you could hope for would be less then 5 position improvement. If the point of this is to increase your chances of winning then the goal must be to be seeded 1-8. The only teams that can possibly do this are seeds 9 though 13. So if every team on the alliance were in this range it would then make sense to pursue the 6v0 strategy.

I cannot see anyone gaining more then 5 seed positions from the most successful execution of 6v0. Of what value is this if you are ranked 25 or higher. At the start of ally selection you are sitting in the same place as the worst seed, on the sidelines waiting and hoping to be selected. I doubt any team even looks at seed numbers when they choose allies. You choose based on your scouting reports.

Someone please explain to me how this is a useful strategy?

 

[Integral]

I have done further analysis of the 6v0 strategy and have found the fatal flaw. Looking at the final seeding the top seeds seeds all have a lot of coopertician points. coopertician points are 2x the LOSSERS score awarded to the winner of the match. It seems to me the optimal stradegy is No Defence each team scores as much as they can with no interference. If a team begins to get to big a lead it is in their favor to score a few points for the other team. Each ball they kick in is worth 2 seed points.

This has the effect that the scores will be close to the end which means anyone can win, given a few cleaver moves at the last few seconds.

http://www.chiefdelphi.com/forums/showthread.php?t=83911&page=8" is a thread that I express myself in at Chief Delphi

 

[Chi Meson]

Our goal is to hang every event. That's it.

If win or lose our heats, or "win-lose," or whatever, if we can advertise "8 out of 8" elevations, I think we will be picked for the finals.

We have a good hanger.

 

[Chi Meson]

I have done further analysis of the 6v0 strategy and have found the fatal flaw. Looking at the final seeding the top seeds seeds all have a lot of coopertician points. coopertician points are 2x the LOSSERS score awarded to the winner of the match. It seems to me the optimal stradegy is No Defence each team scores as much as they can with no interference. If a team begins to get to big a lead it is in their favor to score a few points for the other team. Each ball they kick in is worth 2 seed points.

This has the effect that the scores will be close to the end which means anyone can win, given a few cleaver moves at the last few seconds.

http://www.chiefdelphi.com/forums/showthread.php?t=83911&page=8" is a thread that I express myself in at Chief Delphi

Integral,

After Update 16, do you still think the "no defense" strategy is worthwhile?

 

[Integral]

Absolutely, It means that a "winning" team can hold a larger leading margin. But it remains true that the best possible outcome for a win is a score;

x to x-1 .

The winning alliance gets seed points:

x + 2(x -1) +5 = 3x+3

The losing alliance gets:

x

Before the rule change the optimal outcome was a tie. for a x-x tie ALL teams get 3x seed points. I am hoping a clarification will come out which assigns perhaps a 4pt bonus to each team involved in a tie. Penalties on the winning team will reduce their seed points.

 

[Chi Meson]

I just went to see the WPI (Worchester) Regional. Our team was supposed to do this as our initial competition, but we didn't have the money in time :(

This is what I'm seeing so far:

Autonomous is 15 seconds of mostly dead-time. Occasionally, a team in zone 1 will push their ball in, and a single point is scored.

Nearly all points are scored from "ramming." Scoring from zones 2 or 3 are virtually unheard of.

"Elevating" the robot, when successful, wins the game (more often than not). "Suspending" the robot is unheard of.

There was one team, 230, that successfully "elevated" 6 out of 8 of the preliminaries. They also ended up top seed. Another team that elevated 5 out of 8 ended up fourth seed.

Anyone else with Week 2 observations?

 

[Integral]

Must be nice to have multiple competitions near enough that you can just go to watch.

I have been poring over the results of the completed regionals. It is consistent that the top earners of coopriticitian points are also the top seeds. I am observing that as time passes there are more and more close matches. Though there are still way to many lop sided victories.

Had I to do over again I would have the drive team visit all teams participating in our matches and attempt to convince them to play for a no defense high point tie. If 1 alliance starts to pull ahead then both teams should work to even up the match.

With the 5 pt win bonus an alliance will get more seed points with a 1 or 2 pt differential then for a tie. So perhaps later matches should be kept close then play a "may the best man win" finish.

At Portland hanging was insignificant. Only 2 teams succeeded in scoring a hang. I noticed that in other regionals I observed that only a few had more then a couple of teams hanging. Since I never saw an odd number of hang pts I must assume that no one has suspended yet.

 

[ideasrule]

Autonomous is 15 seconds of mostly dead-time. Occasionally, a team in zone 1 will push their ball in, and a single point is scored.

This makes me ambitious. Our robot should theoretically be capable of scoring from the mid zone, so I can't wait to see whether I can get the robot scoring 2 balls during autonomous.

 

[Chi Meson]

Today begins week 3.

Anyone doing?

We have another two agonizing weeks to wait. I've been looking at our hanging hooks, wishing that we didn't pack up the rest of the hanging mechanism in the crate.

We have spent the last two weeks building the practice robot, re-fabricating the kicker.

The problem is, our practice bot is a two-wheel drive, while our competition bot is four-wheel.

We are still arguing over tank drive vs. single stick drive. My observation was that tank-drive robots tend to be poorly controlled. I made a unilateral executive decision last night: "No tank drive!" That was a thrill!

 

[ideasrule]

Why not program the joysticks so that the driver can switch between tank and arcade any time they want?

 

[Chi Meson]

Simple answer, we only have two joysticks. Although if someone were to buy another one for the team, then that's a possibility/