Week 4: So how's that Field Working?

[Jack Jones]

Please read Dr. Joe's intro FIRST:

Greetings from Milwaukee.

THE GOOD:
Put the Wisconsin Regional down on your "A" list as an event to attend in the future. They've done a great job in planning. nice venue - large pits - smooth flow - on time - great matches.

THE BAD:
It's a FIRST event -> It's all good.

THE UGLY:
The automatic scoring system.

[sanddrag]

I noticed a few scores up in the 300 or so range on the realtime scoring on the video screen. They took it off for a couple matches. Team numbers on the alliance stations have been messed up a few times and the emcee called them the wrong number.

For corner goal incursion, they are giving a DQ and a disable, which is quite interesting. Doesn't hurt the team any more but it does hurt the alliance more.

Other than that is seemed okay.

[Paul Copioli]

Nope. I didn't call the wrong number. The numbers on the screen are wrong and I am looking at the teams' badges to get the team number. Basically, they are not setting the new team number until after we announce the teams.

[Jack Jones]

Originally Posted by Paul Copioli

Nope. I didn't call the wrong number. The numbers on the screen are wrong and I am looking at the teams' badges to get the team number. Basically, they are not setting the new team number until after we announce the teams.

You were in LA too, Paul? Man! You shure get around!

[jparkteach]

Here in NYC the automatic ball counting system is working very well. The only times there were problems was when there was human error (i.e. coaches putting their hands in front of the camera, human player forgetting to remove balls from the coral, teams driving their robots into the corner goals). They have two people (one ref, one real-time scorer) counting every goal just to make sure.

[Ryan M.]

The BMR was surprisingly good after seeing so many other regionals having problems... the only major differences that I noted can from penalties, which obviously aren't counted by the automatic.

[Jack Jones]

Originally Posted by Ryan M.

The BMR was surprisingly good after seeing so many other regionals having problems... the only major differences that I noted can from penalties, which obviously aren't counted by the automatic.

The auto scores in the data were adjusted to account for penalty and ramp points.

surprisingly good

Show us your numbers.

They have two people (one ref, one real-time scorer) counting every goal just to make sure.

Which do they believe? And, show us your numbers.

If the auto scorer gets the autonomous winner wrong - does anybody catch on? If they do, and correct the ten points after the match, how do they change the fact that both teams played the game based on false information?

[Ken Leung]

Here at the Davis regional, we have 6 additional volunteers in ref shirts each in charged of a goal on the field, whose sole role is to count the balls going into them every round.

They each have a white board and marker on their hand, and they write down the number of balls that went into the goal when autonomous mode and the match is over. And each time, our head ref, Stu Bloom, standing in front of the scoring table, have the ball counters hold up their white board to show him what their number is, so he can record them on the little sheet and hands to the scorekeeper to calculate the official score.

So far all day yesterday this system has been working out really well for us. The ball counters are amazing, and they never shift their focus away from the goals. Stu did a great job organizing all these!

[Don Wright]

The main problem is not the final score. That is easily corrected. It's when the autonomous mode is incorrectly scored and the wrong team is declared the winner and goes on defense first.

This has happend to us twice...both in regional finals. One time we were declared the winner and one time the opposing alliance was. Both times were wrong and both times the wrong team went on defense first.

Sure, the 10 point bonus is easy to correct. But what about the change in strategy a team would do based on the outcome of autonomous mode?

The inability of the system to have humans override the autonomous mode scoring to get the winner right is something I hope gets corrected before Atlanta.

[petek]

Originally Posted by Jack Jones

The automatic scoring system.

If I read this right, you're saying the wrong alliance won about 1 out of 5 matches? As an FTA (NJ, Pitts & Philly) I find this very worrying. May I ask where the data comes from?

[Tomasz Bania]

Originally Posted by Don Wright

The main problem is not the final score. That is easily corrected. It's when the autonomous mode is incorrectly scored and the wrong team is declared the winner and goes on defense first.

This has happend to us twice...both in regional finals. One time we were declared the winner and one time the opposing alliance was. Both times were wrong and both times the wrong team went on defense first.
Sure, the 10 point bonus is easy to correct. But what about the change in strategy a team would do based on the outcome of autonomous mode?

The inability of the system to have humans override the autonomous mode scoring to get the winner right is something I hope gets corrected before Atlanta.

Yes, I clearly remember that during Final 2, but one thing still makes me unsure...

Since the balls are sometimes travelling VERY quickly through the goals,
are the Counters 100% dependable? I know they are very good, but sometimes the outcome of the match (F 1-2 in Detroit) would have made a possible difference and a possible win if the autonomous was in our favor.

I don't doubt the counters accuracy, I just don't like the effects of not having 100% accuracy.

Tomasz Bania

[Don Wright]

Originally Posted by Tomasz Bania

Since the balls are sometimes travelling VERY quickly through the goals, are the Counters 100% dependable?

Balls aren't counted as they go through the goal, only as they go down the tube into the can... Hence, if balls bounce out, they aren't counted.

[Tomasz Bania]

Originally Posted by Don Wright

Balls aren't counted as they go through the goal, only as they go down the tube into the can... Hence, if balls bounce out, they aren't counted.

Sorry about that, I men't to say that balls going though the tube sometimes fall very quickly as seen when we both clogged the goals. But I still believe this is a possible, but not very likely, factor. But My argument still stands on corner goals.

Tomasz Bania

[Commanchetruck]

The scoring at Granite State was terrible. When a robot got too close to a coner goal the scoring went way up. I think it had something to do with shadows. FIRST added refs to the corner goals but it didnt really help. During a match we broke down and only one of our robots got to the corner goal. The team started with 10 balls and picked up two on the way over and dumped them. One ball did not go in. The scoring system showed 9. At the end of the match the score was 9-9. We talked to the head ref who talked to the ref at the corner goal. He came back to us and said that the corner ref couldn't be shure how many balls went in. So we ended up with a tie instead of a win. Another time we were with two other shooters. At the 4th period one of the coaches said we were up by 40 and to not shoot anymore and just go on defense. It turns out the board scored wrong and thats where the coach had looked. We ended up losing the match where we could have won.

[Jack Jones]

Originally Posted by petek

If I read this right, you're saying the wrong alliance won about 1 out of 5 matches? As an FTA (NJ, Pitts & Philly) I find this very worrying. May I ask where the data comes from?

Data is from the official score sheet - the standby, hand entered, hard copy.

NO - what it says is...
Assuming the human count was accurate, and had the auto-count been used as the official score, then the wrong alliance would have won 20 percent of the time. What it also say is that the wrong team went on offense after autonomy 1 of 5 times. There was no way to override it.