Robot Diversity

In my opinion, the quality of the game created by the GDC is directly reflected by the diversity of the robots that tackle it. In 2007, diversity ran very high due to the un-forseen nature of the game. Overdrive had slightly less diversity. Then Lunacy came along and had a roughly 4-type system: dumpers, shooters, shumpers, collectors. This year’s game, Breakaway, has about the same amount of diversity or even less with 4 types: kickers, shovers, loopers, and mainly-hangers. Loopers and mainly hangers are rare types, which makes this even less diverse.

My question: Do you think that first has done enough to force creativity, “thinking outside the box”, and diversity among robots, or should they do more?

In an attempt to popularize has the GDC given up some of the awesome design differences that really can inspire kids and force progress.

What are your thoughts, comments, anything on the subject?

I still see a lot of diversity in implementations. This is hidden by the bumpers. It makes every single robot look similar in my opinion.

The GDC has quite a hard job. They have to create a game that is visually appealing, and has different levels of difficulty built within the game. This allows the rookies all the way up to the veterans from 1992 to have fun and enjoy the game.

Yes, this years game may seem boring in terms of diversity as many robots look like boxes, but have you had a chance to see all the robots up close? I think if you were to look at every single “box bot” you would be surprised at how diverse robots this year really are.

Some examples of this: kicker style (motor, pneumatic, spring, electromagnetic, etc), drive train choice, hanging method, etc.

This game also posed several design challenges that you might be overlooking as well. Want to go under the tunnel? You had better be shorter than 17" or it isn’t happening. Want to traverse the bumps? Hope you shock mounted everything.

There are so many more things that are hidden within this years game that most people just overlooked. Strategy wise, yes, there are “few”, but this all changes when you place in things like the ranking system for quals.

I applaud the GDC for this year’s game. It made me personally think even harder this year about design trade offs, something that happens in the real engineering world.

You make good points however those choices about drive train and power are made no matter what the game is. An example of diversity I would bring up would be team 148’s nonagon in overdrive. That thing was outside the box. 469 was this year too, but most other teams are similar in that they have a way of possession and a way to kick and a tank drive. Thats the problem i think, that robots look and do the same things. Looping was a cool, unique idea, that managed to conquer the game.

Yes, most of the robots are hidden by the bumpers this year. however, despite mechanical differences sucessful the robots for the most all kick the same, grab the ball basically the same (double roller pincher) and if they can hang the same way (vertical bar). Of course, their are a lot of exceptions to the second two, but its really hard to tell the difference between kickers.

I think that a lot of this is due to the vast amounts of rules limiting what you can and cannot do. For example how different do you think the designs would be if there was no three inch rule? Just changing that would increase the robot spectrum.

Perhaps, next year the GDC should simply make the game a lot harder (at least harder then pushing balls into a goal), but not limit teams in what they can build to compete (robot out of the box extention, touching or going to certian parts of the field at certian times, interaction with game objects, ect. I think everyone would be happy with this plus no lawyering of the rules. That way if you can dream it you can build it.:slight_smile:

Bjc I agree with what you’re saying, almost everyone is either winch, cam, or pneumatically kicking and almost every successful team uses a pincher or roller. Most teams are switching to the grab and turn upward bar due to speed and drive trains are almost all tank. I am not trying to insult anyone’s designing if they use these things, I mean my team does too, this year in my opinion is about how good your driver is and how smoothly you do the same things as everyone else (with the exception of loops). Simbotics, I think, does these things more smoothly than most and has a good driver, these are the reasons for their success.

I think the way to encourage diversity in a game is to provide multiple, unrelated methods of scoring and allow teams to choose which method they wish to focus on.

In my opinion, the GDC has done a great job of this in past years. In 2006, you had high goals and low goals. In 2007, you had ringers and the elevation bonus. 2008 was excellent! You could hurdle, herd, cross lines, or place to get points in teleop, and autonomous had its own special scoring. 2009 was probably the worst year I’ve seen in this regard: shooting moonrocks into baskets was the only way to score, and the bonus scoring methods (e-cells and supercells) required pretty much the same engineering as the regular scoring method. Having said that, 2009 (my high school senior year) is the year I learned the most from FIRST, because the diversity was still there; it had just been largely moved into software (traction control).

In 2010, you have goals, elevation, and suspension as scoring methods. The cost/benefit ratio for suspension is so low, I doubt we’ll see it more than once, so that one doesn’t really count. So I would say the diversity of this game, with two main scoring methods, is about average.

I rate diversity as great bots (think 1114), good bots, ok bots, not so good bots, bad bots, should not have passed inspection bots (waaaaaay too many this year).
I’d rather rate games on the challenge (this game has vexed many good veteran teams) and how exciting the elims are (this year has all been pulse pounding. Lat year? It was like watching snails joust).

I actually disagree with you, the veteran teams and previous year’s winners are dominating this year, absolutely. 2007 saw less veteran and usual winners doing well which is why i loved it. Think 1114, 217, 67, 148 and so on.

Even if robots have all the same abilities or most of the robots have similar tactics, there is also how that tactic is implemented during building and how effective it is. While the outside of robots can be very different, the inner workings makes every robot different, and slight differences that you can’t see on the outside can change everything.

Also, I originally disliked games where robot diversity was at a low, but especially this year I don’t seem to mind it at all. I think this year, having a bunch of robots that are very similar really focuses on how that team plays the game with the robot they have, which is a good thing…but I agree, sometimes you sometimes what to see a big shift in styles and play styles.

You’re going to have to broaden your perspective in order to see all the diversity. Going by your logic of assigning robot types simple labels, I’d say that 2007’s game was even less diverse than this year’s game. There were three main types of robots in '07: Ramp-bots, Ring-bots, and Ramp+Ring bots. But… if you wanted to look into those three categories, you’d see amazing feats of engineering and diversity. There were pneumatic ramps, two-jointed arms, four-bar linkages, etc.

Take a look at this year’s robots. On the outside, they may all seem extremely similar, which I personally think is due to the bumpers this year (yet some teams like 1501prove me wrong :p). They all may look like red and blue squares and rectangles on the outside, but you have to look a little deeper. Most 'bots have a possessing mechanism of some kind, whether it be a pincher-roller set-up, a single roller, or a vacuum. (Hey, that’s three types right there!)

Look at the hanging mechanisms this year… There are some teams that drive up the vertical bar, there are some that grip and flip themselves over the vertical bar, there are some that grab the top of the tower, etc. Not only that, but the means that these robots accomplish these feats: scissor lifts, winches, pneumatics, gas shocks…

Look at all the drive trains this year… There are, of course, your standard four wheel drive, six wheel drive, and swerve drive (dare I call a swerve drive “standard”? :ahh: It is becoming increasingly common, however.) But, have you ever seen so many Mecanum drives attempted by so many bots? How about eight wheel drives? What about articulating eight wheel drives? And then there’s that non-a-drive. :stuck_out_tongue:

Now… don’t you even get me started on 469. Heh.

TL;DR: Are the robots similar this year (And most years, for that matter)? In a very broad sense… yes. But if you look a little closer… no.

Yes, it is about how smoothly you do things. But every year always relies on driver skill. And you said Simbotics does things smoothly. I pray that you are talking about the design of it and not how smoothly their driver moves the robot… Because no ones driver is worth crap if their chains act funky and the kicker jerks the whole robot when it kicks.

Just my $0.01

-Rion

@formerly famous
I was definitely talking about the smoothness of their robot, not necessarily their driver.

This was the game that broke Pink’s streak and 111 didn’t come away with a banner like they usually do.
This game is tough.

Excellent point, some teams were beaten by it, but i think the general trend is leaning towards veterans, that could be just me.

Alot of veterans took to this game like a duck to water.
But those teams that you mentioned do that year after year after year.

I actually think this may be the best aspect of the game. Anyone that can build a driveable robot can score. When was the last time a rookie team with a kitbot basically do the same thing as insert your favorite powerhouse with their professionally machined robot? In 2006, a kit bot could only herd balls into the goal, which was 1/3 of the goals in the upper goal. In 2007, a kit bot was forced to play defense and climb ramps. In 2008, a kit bot could only do laps, which was 1/10th the points of a hurdle and lap. In 2009, a kit bot just had to prevent being scored on. But, this year a kit bot can do the main scoring objective.

But, better yet, well designed a built robots are still rewarded! A team with a good ball magnet and kicker can be super more effective in the close zone. Also, teams with kickers are necessary for an alliance to get balls to the close zone. To give a further advantage to the overachieving teams, they can hang for a 2 point bonus.

To sum it up, this year is good because anyone can score, but those teams that build great robots still can excell.

Your point about everyone being able to score is interesting, I never thought of it that way, but this reason is why I think the diversity is not as high.

Sometime veteran teams just have ‘not quite as good’ seasons. For example last year team 33 built a turreted shooter instead of the much more effective dumper and we just couldn’t find our groove.

This year, however, I think we hit the nail on the head with our robot design:) .

Coming up with a design for the competition and actually engineering that design are very different things, but both require creativity to accomplish. The GDC I think has given us plenty to consider, with the bump, the 3 inch rule, the tower, etc. Many teams have come up with designs that look similar, but a lot of them are really quite different. You may have a ball roller or a vacuum, and an arm that winches itself up the tower, lifts itself off the floor by making the robot expand downwards, grabs the tower and rotates the robot’s frame so that it is above the underpass, etc.

FIRST has come up with a very exciting competition this year, and while many robots are more similar than they are different, the design challenges of this year have pushed a lot of teams to their very limits. Even if robots are not so original, there are many new things teams had to consider this year, like how to use the camera, and strategies centering around this year’s scoring system. Deep down, I think as much, or even a lot more thought went into many robots this year than in the previous two.