Week 0 Discussion

[Ginger Power]

Post any reactions to your week 0 events here. I'd love to see discussion about the game, how it's played, how your robot does, etc.

I'll be attending the Centurion-Krawler event on Saturday, and the Chanhassen week 0 event on Sunday. Couldn't have a better birthday present than to watch robots all day!

[Steve Horn]

In Suffield, CT Aces High is hosting Suffield Shakedown today. To watch the webcast go to http://www.shsrobotics.org/ and click on the "Watch Now" link. Good luck to everyone hosting events today and teams that are attending the events.

[Foster]

Looks like it was lots of fun. Looks like destroying the defenses was literal, lots of wood chunks flying around.

Lots of interesting drive trains, it was nice to see the Falcons "10 wheel drive" in action.

Lots and lots of low bar robots today, only saw a few that had to go other directions.

Really liked how you let teams play for 5 minutes a time. It took drivers about that long to figure things out.

Good event, this looks like it will be a good game to watch from the stands.

[cjl2625]

Aiming is hard.

[Rangel(kf7fdb)]

Originally Posted by cjl2625

Aiming is hard.

What I saw at the Arizona duel in the desert as well. It seems my assumption that any shooting when not on the batter will require a solid vision tracking system. The margin of error is just too low for a driver to consistently aim manually. Maybe I'll be wrong come championships but so far it seems to be the case.

[wjd13]

I was most surprised to see that not one team at the Merrimack event managed to damage a Category C defense on their own. They either needed 2 robots, or spent the whole match trying to get across once. Did anyone at the other week 0 events manage to solo a Category C defense?

[IronicDeadBird]

Originally Posted by Rangel(kf7fdb)

What I saw at the Arizona duel in the desert as well. It seems my assumption that any shooting when not on the batter will require a solid vision tracking system. The margin of error is just too low for a driver to consistently aim manually. Maybe I'll be wrong come championships but so far it seems to be the case.

Was aiming hard due to just the inherent difficulty of shooting from farther away/lack of alignment or was it due to defensive driving?

[Rangel(kf7fdb)]

Originally Posted by IronicDeadBird

Was aiming hard due to just the inherent difficulty of shooting from farther away/lack of alignment or was it due to defensive driving?

We weren't personally practicing aiming but from what I saw of every other team, it was quite a challenge to line up accurately. For one it is really hard to see with the defenses in the way and the margin of error is just very small. Even from the archived webcasts I'm watching from the other week 0 events, aiming is a challenge. This difficulty will be magnified when robots start playing defense.

It's also hard for a lot of teams to turn because of their drive train. Making a high traction drive train to get over defenses is nice, but it also makes it harder to turn efficiently making it harder to aim.

[MrForbes]

We made a minor modification to our robot which helped shooting accuracy. Our spotlight for aiming seems to work OK.

[MecaNaught]

Over here at team 3102's Week Zero the competition itself was exceptional, and we got the process going smoothely to make it great for everyone.

The Problems:

We were using FMS Offseason

Only about half of the OM5P routers there would configure properly, even with firmware flashes. We could take one router, configure it fine, take another, wouldn't work. Even three computers wouldn't do it when it did others fine, then randomly after trying all day it would work. Unreliable at best. We had to borrow out a few older D-links B's to teams, and those configured perfectly every time.

Teams CAN NOT connect to a central router to practice with their robots. Teams were controlling other teams and seeing other cameras, and we made sure to re-configure their routers to their proper team numbers. I was told that the DNS method implimented in FRC's software does that natively. With FMS software enabled we could get two robots working correctly but didn't have time to do more testing. This ruined our entire FMS setup because there was no time to take everyones radio's and re-configure them, especially not with the configuration problems.

The solutions:

Teams were told to reconfigure their own routers with the config software, so we had about 11 seperate AP's going. Much interference with our stream was had, however teams could operate properly.


We simply had teams use the 'practice' method and set their timing to:
0
15
0
115
20

I had to do a good 30 laps around the field overall, showing teams before every round the timing and to press enable practice when we do a ready check and count down.

Next Year:

Next year if the same issues persist we will simply inform teams to configure their own radios, then right before lunch we'l gather them up and configure them while lunch goes on. Then we should be able to control them via FMS for official rounds.

We will also be running a dedicated line to have a consistent live stream, another camera may be added.

Thank you to those that came

We greatly appreciated the crowd we drew. Solving problems and getting everyone operational was a tedious, but very rewarding task. Its really great seeing all of you compete on the field and testing out your equipment. Sorry about the music, nobody thought to hire a DJ This is our first year with our official Andymark field, next year we'l try to make it all just right.

[team-4480]

Originally Posted by MecaNaught

Thank you to those that came

We had a really fun time at your event! It was our first time doing a Week Zero event and we absolutely loved it. All of your volunteers were super helpful!

At the event, our we found issues with:

- Ramparts! They seem to be the devil in disguise! They look so easy until it turns your 70 degrees into the next defense!

- As others have said, the high goal is super hard to line up. We didn't even have a defender on us either. I will definitely have to work on some vision if we want to be more consistent.

-Speaking of vision, at the event, I could not find one team that was using vision. There was about 12 teams there and I thought I would find at least one team with a good vision system.

- The Portcullis, Sally Port, and Drawbridge were neglected as if they were poisonous or something.

- The Spy Human player was actually super useful for lining up shots when obstacles were in the way. I thought the Spy was going to be useless but was proven quite wrong.

- We need some more sensors so our drivers can be more informed on whether or not the ball is in and how far we are away we are.

- I think we should pick only low defenses when possible because the Driver's vision was super annoying. We occasionally accidently hit another friendly robot because the big drawbridge was in the way. Cameras can only do you so good....

[KrazyCarl92]

I was watching the action at Suffield Shakedown briefly today. I noticed that at least one set of the ramparts used was actually backwards. Just wanted to throw that out there for any teams relying on the scrimmage for their testing. Our team definitely noticed a significant impact that the direction of the ramparts (I.e. which side is up/down) has on how it should be crossed.

Particularly if you plan on crossing this one in auto, it should be done with the proper direction of the ramps.

[cjl2625]

Originally Posted by IronicDeadBird

Was aiming hard due to just the inherent difficulty of shooting from farther away/lack of alignment or was it due to defensive driving?

Oh god, under defense, high goal from the middle of the field will be a nightmare. Even if I write vision code, by the time we're aligned, a defender will have pushed us half way across the field.

[pipsqueaker]

Originally Posted by cjl2625

Oh god, under defense, high goal from the middle of the field will be a nightmare. Even if I write vision code, by the time we're aligned, a defender will have pushed us half way across the field.

I have to ask, how did you guys come up with such a complicated shooter?

[cjl2625]

Originally Posted by pipsqueaker

I have to ask, how did you guys come up with such a complicated shooter?

Good question, I'm not entirely sure. Though I do know a few factors that led up to the design. For example, the linear shooter (one cim, two sets of wheels on one side to get some ball spin) seemed the most promising design during prototyping, we wanted the harvester and shooter to be separate mechanisms and face opposite sides, and we wanted the release point of the ball not to be too low. So from there we came up with a linear shooter assembly on an arm.