Game Strategy

[gbf1991]

So I was wondering what sort of strategies other teams were going with. Our team is creating a bot that can kick, herd, go through the tunnel, go over the bumps, hang from the tower, and climb on the platform. We hope to fill in any vacancies on our alliance by being able to do almost everything in almost every zone.

[BrendanB]

So I was wondering what sort of strategies other teams were going with. Our team is creating a bot that can kick, herd, go through the tunnel, go over the bumps, hang from the tower, and climb on the platform. We hope to fill in any vacancies on our alliance by being able to do almost everything in almost every zone.
Wow! That is quite a robot you have! We are planning on scoring in the goals and some other stuff, but I'll wait a little longer to release them! :]

[Daniel_LaFleur]

So I was wondering what sort of strategies other teams were going with. Our team is creating a bot that can kick, herd, go through the tunnel, go over the bumps, hang from the tower, and climb on the platform. We hope to fill in any vacancies on our alliance by being able to do almost everything in almost every zone.

Until the BAE unveiling, we're going to keep our 'bot under wraps (Besides the teasers :P ).

Once that happends, we'll give out all of our robots spec's and it's capabilities.

[sloteera]

Our bot is being build for play in the FAR ZONE, and our thought is "to defend" the ball instead the robot. We pretend to kick every ball in the FAR ZONE to the NEAR ZONE!

[cmh0114]

So I was wondering what sort of strategies other teams were going with. Our team is creating a bot that can kick, herd, go through the tunnel, go over the bumps, hang from the tower, and climb on the platform. We hope to fill in any vacancies on our alliance by being able to do almost everything in almost every zone.

That sounds like quite a robot. How are you planning on going through the tunnel and go over the bump? It seems like that would take a lot of over-complicated engineering.

[dodar]

That sounds like quite a robot. How are you planning on going through the tunnel and go over the bump? It seems like that would take a lot of over-complicated engineering.

Thats what we are doing too, we are going through the tunnel and over the burm

[BryantWebb]

Our robot will be able to both go over the bump and under the tunnel

[gblake]

So I was wondering what sort of strategies other teams were going with. Our team is creating a bot that can kick, herd, go through the tunnel, go over the bumps, hang from the tower, and climb on the platform. We hope to fill in any vacancies on our alliance by being able to do almost everything in almost every zone.gbf - That's an interesting list of Functions, but it isn't a Strategy. A strategy tells us how, in appropriate situations, you will use them to gain an advantage over opponents and how you will use them to counter opponents' strengths. You and your team may have another chunk of work (thinking and practicing) ahead of you.

With so many functions, I'm very curious about how you think they can best be used during the 135 seconds of a match. Have you determined that each one of them will be required in some likely situation? If you were to describe those situations, I'll bet we could get more than one good discussion going.

During a match your driver(s) will need to be able to recognize/anticipate the situations your robot is designed for, and then use the right features to succeed in that situation. With all of your robot's tools at his disposal, your driver(s) will need extensive practice/training.

Good luck on the field,
Blake
PS: This same hint applies to everyone who listed robot capabilities/functions, but has not yet decided in which situations to use each of those capabilities.

[BrendanB]

That sounds like quite a robot. How are you planning on going through the tunnel and go over the bump? It seems like that would take a lot of over-complicated engineering.

It is actually not that hard to do it all. Large wheels, short base, compact elevation mechanism, and a small kicker.

Not that we don't have experience in some of these designs!;)

[Taylor]

"You must spread some reputation around before giving it to gblake again."

At this point in the build season, I think many teams would be happy to discuss their robots' abilities, but I don't think I'm alone in saying I would rather not discuss our team's strategies. The reason for this is threefold - strategies can be very situation-specific, all strategies can be countered, and our gameplay strategies are only to be shared with our alliance partners.

[Chris is me]

The debate over strategic flexibility versus a specifically toned game aspect is one that defines most teams, seasons, and games. Doing everything perfectly, without compromise, is the ideal, but that's rarely possible for even the best of the best.

So then you decide if you do everything with compromise, some things without compromise, or one thing perfectly. In games with alliances, rarely does the need or option arise for one robot to be able to do everything it takes to win a match; the GDC doesn't try to design alliance-based games that are winnable by one. So the general opinion of me is that it's better to make less functions perfect than more functions work.

That doesn't mean if you can't perfect something not to do it though. For example, 2791 has bump climbing ability. It is not perfect or a big focus of the robot. It is an option on our robot because we feel at least one alliance member will always need to be able to traverse them, so the best way to guarantee that is to make us that alliance member. Not focusing on making a crazy suspension or other, more effective bump traversing techniques lets us put more time into the kicker, while having the bump climbing option as designed now doesn't interfere with the kicker's success, as it took fairly little effort to design the drive to work over the bump relative to the option we would have chosen without the bump in mind. This is highly variable depending on your team's resources.


So that's my thoughts on "listing what your robot can do", really. I don't care what you can do. I care what you can do well and consistently.

------------

As for "game strategy", this year's game mandates a few things, in my personal opinion:

Ball return control
Flexibility / Transition game (offense to defense)
Ball hoarding in offensive zones

The ball return is pretty huge. With only 12 game pieces, every game piece will be fought for in the middle zone. An alliance will need to have a consistent middle ground opponent with a high traction drive (sorry, mecanum probably won't cut it here). The same limited number of balls means that the game's focus will change based on where said balls are. When your opponent has more balls than you do in their zone, you'll have to pull back and play defense, but ideally there will also be times when your zone has the balls, so an alliance member will need to be able to go from the opponent's zone to the middle to the front in order to play where the balls are. The unique one robot in opponent's zone rule means that balls in your own zone are more "safe" than anywhere else on the field, so almost secondary to scoring points is securing balls in said zone.

From this strategy, you'll get the design criteria for 2791's robot. The first aspect mandates either skid steer or possibly swerve drive for the traction and pushing power. The second, combined with the design expertise of the team and the 6 week period, requires a bump climbing skid skeer. The third requires a kicker that can shoot over at least one, preferably two bumps, even at the expense of a little accuracy.

The real tricky part is being able to adapt your strategy to your alliance. The only constant is yourself, so you have to make your robot what you believe is absolutely essential to a winning alliance to ensure the most wins possible. This varies from team to team, obviously; you can quite clearly see that a lot of teams have decided this is the year for mecanum drive! However, it's best to put any worrying over what you can't control off until after build. Once you know yourself, then consider how you will tackle qualifying matches where your robot is the only one with a kicker, or you have the only traction drive, etc.

Another thing to keep in mind is that you're probably wrong. I bet Week 1, I'll look at this post and realize I either overemphasized or overlooked some game aspect I wish I hadn't. Pay attention to the regionals before you and see if your strategy would work. Consider adapting if it would not.

All this being said, I'm sure if all goes well, I'll regret posting this. Use it as you wish.

[Bill_B]

"You must spread some reputation around before giving it to gblake again."

At this point in the build season, I think many teams would be happy to discuss their robots' abilities, but I don't think I'm alone in saying I would rather not discuss our team's strategies. The reason for this is threefold - strategies can be very situation-specific, all strategies can be countered, and our gameplay strategies are only to be shared with our alliance partners.

So, should we start another thread in which we list and describe the situations that we expect to see? I've been trying to get my team interested in making a table/chart/list of situations for a long time now. Maybe I'm not looking hard enough, but I don't see a great deal of interest in discussing the game strategy (or TACTICS) among the members of the team. Perhaps they think they can do all that the morning of their first competition?

[gblake]

So, should we start another thread in which we list and describe the situations that we expect to see? I've been trying to get my team interested in making a table/chart/list of situations for a long time now. Maybe I'm not looking hard enough, but I don't see a great deal of interest in discussing the game strategy (or TACTICS) among the members of the team. Perhaps they think they can do all that the morning of their first competition?Ask your team to play a game of football/soccer/basketball/lacrosse. I suspect that you will get a fair amount of interest and enthusiasm. Ask you team to map out and practice plays to execute in various game situations for one of those sports, and I suspect you will get considerably less interest.

However, to get past the level of backyard or parking lot pick-up game teamwork, you must either do that (think and plan for a while), or practice and play together for a long time. Most teams aren't going to get a ton of practice with their real FIRST robot; and you certainly aren't going to get any non-trivial practice time with your qualification round allies. So, you need an effective alternative.

Consider having the students imagine they are on a field (use tape or traffic cones or chairs or whatever to mark some boundaries in a large space somewhere). Let 6 be the pretend robots and 6-12 more be the drivers. Toss a few soccer balls into the imaginary field and have the drivers start yelling instructions to the robots. Chaos is likely to ensue. Have some fun and also start bringing order to the chaos. Slip some strategic and tactical planning into the fun.

Or... Download a Breakaway simulator (I think two are advertised in Chief Delphi threads this year) and use computers to do the same thing. Multiple players and balls on the field at various locations, what do you do? Shoot, pass, defend, play keep away, hoard, move to another part of the field, go over the bump or through the tunnel, prepare for the end game, score in the other alliances goal?

At least one of the simulators lets you replay matches so that you review the game films and discuss whether players made good choices or if they missed opportunities to do better.

Initially, everyone wants to play :), and no one wants to study :rolleyes:. So turn play into studying. ::safety::

Blake