why sooo many bad robots in 07

[Joel J]

I haven't seen alot of bad robots that were bad because they tried to do everything, though I do not deny their existence. I've seen alot of bad robots because of a poor gripper, or a poor drivetrain. Usually, it has been a poor gripper. By poor gripper I mean slow grab, poor hold, slow release, awkward release onto the rack, holding the tube vertically or worse, etc. Also, there are a few that attempted to be ramps only, but failed, or ended with sub-par ramps.

I agree that there are alot of bad robots, but I disagree that there are more bad robots than normal.

[AznPrincess3089]

Originally Posted by Joel J.

I haven't seen alot of bad robots that were bad because they tried to do everything, though I do not deny their existence. I've seen alot of bad robots because of a poor gripper, or a poor drivetrain. Usually, it has been a poor gripper. By poor gripper I mean slow grab, slow release, awkward release onto the rack, holding the tube vertically, etc. Also, there are a few that attempted to be ramps only, but failed, or ended with sub-par ramps.

I agree that there are alot of bad robots, I disagree that there are more bad robots than normal.

I admit that, at first, my team did try to do arm and ramp but it was in the middle of the building process when we realized that it just wasnt happening so we became a SUPER COOL rampbot.

[Bongle]

I'm not so sure there are that many 'bad' ones as compared to previous years. You being on team 25, you would've gone to championships, so your last memory of FIRST would have been the arbitrarily high quality at Atlanta, rather than the much more down-to-earth quality at regionals. Think to a regional last year: How many robots could shoot autonomous? How many robots couldn't shoot at all? How many teams did you see just setting ramming autonomous modes because they couldn't do any better?

So here's my theory:
I think more teams might be thought of as bad because there is only one way to score points this year within the non-endgame time, and it requires advanced manipulating. Last year, if your shooter didn't work, you could at least ram balls in the low goals. In 2005, you could push tetras under the towers. In 2004, you could, again, push balls into the low goals. In 2003, you could push bins across the playing field into your own zone. There is no way for a box-on-wheels to score this year other than to climb a ramp, and that's more of an end-game bonus that all previous years had anyway.

So the short version: There are just as many bad robots as last year, but this year's game has fewer opportunities for them to make lemonade from their lemons, unlike past games where there was always something 'simple' to do.

[Danny Diaz]

Originally Posted by Bongle

...How many teams did you see just setting ramming autonomous modes because they couldn't do any better?

ROFL. A solid defense is sometimes the best offence - if you can't do any better than the guy on the opposite alliance, but you can stop the other guy from scoring, it's perfectly valid to shut him down. Bumper-on-Bumper contact is very legal.

Watch the fun starting at time mark 1:50.
mms://sargasso-3.arc.nasa.gov/2006/first/lonestar/040106f3.wmv

BTW, we have much respect for team 118, last year at Lone Star they were completely dominant in autonomous. However, being the best of the best does have its downside, you have a lot of people who analyze you very thoroughly looking for a [narrow opening] in the armor. When someone finds it, you should see it as reverence - someone has been thinking about you over the course of 2-3 days more than you've probably thought about yourself!

-Danny

[oddjob]

It looks like quite a few teams have reused drivetrains from previous years, and ended up with a robot that has little ground clearance and can't climb a ramp. Or they clean sheet designed a robot that can't climb. The ramp bonus is so large that this pretty much eliminates those robots from challenging for nationals.

It's an interesting juggle this year, deciding where to best allocate the weight to design a robot that does at least one thing well. I dare say that most teams have to live with decisions made early in the 6 week build window and if the end result is not optimal, time and money constraints limit their ability to make improvements.

[meatmanek]

Originally Posted by Joel J.

I haven't seen alot of bad robots that were bad because they tried to do everything, though I do not deny their existence. I've seen alot of bad robots because of a poor gripper, or a poor drivetrain. Usually, it has been a poor gripper. By poor gripper I mean slow grab, poor hold, slow release, awkward release onto the rack, holding the tube vertically or worse, etc. Also, there are a few that attempted to be ramps only, but failed, or ended with sub-par ramps.

Don't be dissing on the vertical tube handling, man. As soon as there's video of 868's robot online, I'll show you how much vertical tube handling rocks.

But yeah, I'll agree that I've seen a lot of terribly inefficient grabbers. Seems like it takes most robots 10-15 seconds to pick up a tube.
Grabber mistakes I've seen:
2 flat, parallel pincers - The tube flies out when you turn
Grab from the top - You have to aim in 2 directions now
Nothing on the robot to help align the tube - makes the driver aim more
No positional control on the arm (Potentiometers are cheap, people.) - Makes the operator's job incredibly difficult
Lack of tube sensors - Makes the driver react, rather than the robot.

A few sensors and a bit of programming goes a long way. It also helps quite a bit if you plan ahead and design so that sensors are possible.

[T3_1565]

Grab from the top - You have to aim in 2 directions now

well I have seen a few robots that can score this way quite nicely, and thursday will be our test to see if we can too, I'm not too sure what you mean by aiming in two directions, but thankfully we can grab from whatever part of the tube we see fit but our main way is from over top, so our claw doesn't get caught on the spider foot.. we will have to see I guess