FRC Game in 2009

[Doug G]

Originally Posted by Hachiban VIII

Usually when FIRST introduces new technology they expect us to use it.

I guarantee you there will be some mega-bonus for doing something fancy in auton/hybrid with the new controller.

The last change to the controller was 2004 and there was very little autonomous bonus for knocking off the balls on the Tees, all that would happen the balls from above would drop down a few seconds earlier than normal and a 5 pt ball could be put in play (but I never saw it used). Hardly a mega-bonus.

The introduction of the camera in 2005 - upped the ante a bit with the vision tetra - but again, not much advantage and probably the most boring autonomous they've made yet. Only a handful of teams across the nation managed to get the vision tetra, most of the rest of auto-mode capable bots just knocked down the hanging tetra or moved into position. BORING!!

So I'm not predicting a huge mega-bonus for autonomous this year. I think the game will rely heavily on mechanical design and functionality and from the tip on Bill's Blog - the standard frames/bases and drive systems teams are getting used to will most likey need to be modified this year. We've had 4 YEARS of a FLAT FIELD - I bet this year there will be some kind of structure on the field the robots most get over OR get under. Ramp / Stepped Platform / Low Overhead bar or some combination of these. In 2001 they had the teeter totter but on both sides of that they had a 4" high step with a 14" high bar above it. It should be exciting to see what January will bring!!

[lingomaniac88]

I think the game will be something that can be made easier if you take advantage of what the cRIO can do. Being my team's main programmer, I think this will be one interesting year for me.

As a side note, Dave Lavery should be sending another red herring our way any minute now...

[Andrew Schreiber]

Originally Posted by Doug G

The introduction of the camera in 2005 - upped the ante a bit with the vision tetra - but again, not much advantage and probably the most boring autonomous they've made yet. Only a handful of teams across the nation managed to get the vision tetra, most of the rest of auto-mode capable bots just knocked down the hanging tetra or moved into position. BORING!!

Did you see PINK's auton that year? Very little was boring about it, they picked up two tetras autonomously, honestly that and 910's double keeper in 2007 rank among my favorite auton plays of all time

However, I do have to note a trend towards decreasing the value of autonomous. 2006 had quite possibly the most influential auton ever, a lot of matches were won and lost purely on auton. Seeing a year like that would make me happy.But seeing the trend FIRST has been taking towards leveling the playing field I agree that we will most likely NOT see major bonuses for auton this year simply due to the issues most teams will have getting anything to work. I expect the day after we get our control systems CD will be full of people asking how it works. (On this topic, please do not name threads, "Help!!! URGENT!!!!!!1")

[Doug G]

Originally Posted by Andrew Schreiber

Did you see PINK's auton that year? Very little was boring about it, they picked up two tetras autonomously, honestly that and 910's double keeper in 2007 rank among my favorite auton plays of all time

However, I do have to note a trend towards decreasing the value of autonomous. 2006 had quite possibly the most influential auton ever, a lot of matches were won and lost purely on auton.

My Boring Statement had to do with watching 98% of teams do nothing or just simply knocking down a hanging tetra. PINK was in our division at nat's that year and their auto was hands down awesome.

2007 auto was fun, but just didn't get enough value - so often teams didn't even bother.

I think last years auto would be a good predictor for what we might see next year. I mean most teams had their robot at least drive across one line and often two. That in itself was fun to watch. Then there were the Masters class of bots knocking one or more balls and driving a lap. 2006 was similar as well, many teams good come up with a defense drive auto mode or a simple ball dump in the low goal. However if a top team put 8-10 balls in the top goal in auto - they won most of the time. Heck when we put 8-10 balls in the top goal in regular match play, we won most of the time.

[Burmeister #279]

Does that mean that 2008's auton's were the best yet?

[Doug G]

Originally Posted by Burmeister #279

Does that mean that 2008's auton's were the best yet?

This is all somewhat subjective anyhow - but as a fan, my favorite auto modes have been,

1) 2003 - Stack Attack - A race to get to the top of the ramp first and knock the stack over was very fun to watch. Many teams had success and made the first 15 seconds very exciting. In fact with that game the first 15 seconds and the last 15 seconds were the best parts.

2) 2006 - Aim High - So much variety in auto modes. Defensive modes that just drove across the field and hoped to hit another bot. Offensive ball dumpers that put 10 balls in the low goal. And for some they could actually score in the high goal.

3) 2008 - Overdrive - Watching the progression in auto modes in each match was so cool. In the first few matches not a lot, but by the elimination rounds on Saturday, teams could do several lines and a few could knock down a ball. Just seeing almost all the bots participate in the hybrid period was cool.

So I don't rate 2008 as the best - but probably the most simplest one to get something done for almost all the teams.

[ikhana870]

Originally Posted by mikecombes

well we are supposed to have cameras so we started brain storming like this corn maze that we can't see but the crowd can see and we have to drive around using a camera on the robot. (just a thought probally not gunna happen ((THIS year at least lol)))

The problem with this idea is that FIRST didn't give us any type of a monitor to view what the camera is seeing, and I doubt that they expect each team to buy their own.

[Daniel_LaFleur]

Originally Posted by ikhana870

The problem with this idea is that FIRST didn't give us any type of a monitor to view what the camera is seeing, and I doubt that they expect each team to buy their own.

Really? You've seen the 2009 kit and know that there is no monitor in there?

[davidf]

I am new to first, however I have been using LabView for a while at my job and I have a few things to say about it.

1) It is an extremely powerful development environment. Programming in LV may take a while to learn, however once you understand the programming structure, it is much faster to do complex tasks then a similar C or C++ application is.
2) The FRC version of LabView is not, from what I can see, limited, and includes several expensive toolkits including NI Vision.

You are basically limited by hardware constraints and your imagination only. For my work (electrical engineering, RF) I have used labview quite extensively. It would be relatively simple to create a program that searched for objects of various shapes, sizes or colors. It would not be much harder to drive up to them and get a closer picture. Perhaps read a sign and follow its instructions (OCR vi's come in NI Vision that make character recognition straight-forward) or do some complex pattern recognition to sort objects into categories from the field.

Some limitation will be camera focus - I'm not sure if the camera for 2009 supports controlling focus via software - which would be necessary for reading close up text, barcodes, etc.

I know that building a FRC "robot" is challenge enough when it is remotely operated - but I think LabView will make autonomous actions much more feasible and maybe someday FRC will become an all autonomous competition.

For those of you who don't know it: The robot can be programmed remotely via the driver station and a laptop. This also means you can build a frontend on the laptop, e.g. any kind of control station you want on the laptop to give visual feedback or run pre-programmed drive loops or whatever the rules allow. Out of the box you can start a new LV benchtop vi that lets you see what the camera sees and all the analog, digital, pwm, and battery states directly on a laptop.

And perhaps next year when they enable the advanced functionality of the motor controllers-you will be able create feedback loops-and combine this with internal and external sensors (think gyroscopic sensor + accelerometer) and you can have much greater motor control, similar to what is used by Honda and others to build walking, upright android robots.

This kit your receiving is very much a professional, industry standard setup, if you go into electrical engineering / control systems engineering, you can expect to see much of the same software and equipment at your job.

[ikhana870]

Originally Posted by Daniel_LaFleur

Really? You've seen the 2009 kit and know that there is no monitor in there?

you have a point.........

[bobwrit]

Considering the control system is new, I agree that FIRST wont do something radical with the game design this year. I'm getting a feeling that 2010 gonna be a radical design shift. My predicion is it's probaly be some game involving throwing some sort of object.(I know, I'm not being specific here but that's all I have at the monment.)

[ehcualp]

The first hint was posted for what 2009's competition might be:
http://www.usfirst.org/community/frc....aspx?id=11310
its a fish.
could this be hinting a a water robot? a submarine?

[Stealth]

No you just flop around on the course like a fish out of watter

[GGCO]

Notice that the picture is of a fish out of water. Not in the water. What does this mean? Maybe it is just my imagination. Any thoughts?

[Manyu192]

If you look closely at the background, there's a watermarked shape and some text... its hard to read, i think there are a couple of pages stacked on top of each other... the newspaper that fish is wrapped in maybe?

I think the game will work like a fish market, you have to throw slippery fish into buckets...