What can your robot do in Autonomous?

[nehalita]

Just curious what people plan to do in autonomous and what they are capable of.
Are you making more than one auto? Which one do you plan to do the most?

If you plan to shoot in auto, are you using the camera or dead reckoning?

Our team has more than one auto and plans to use the shooter the most.

[Elgin Clock]

I think our plan is to drive forward just a tad, and lock onto the green target and then shoot 10 balls in.

That's the plan anyways. Auto-mode wasn't proven extensively due to time constraints, but when we did, it worked. :)

Actually, we can drive to the floor goal as well and drop balls their too.

So, I guess we have 2 solid ones as of now.

[Rick TYler]

Here is what the software people planned:

1. Drive out a few feet (selectable before start).
2. Turn towards goal.
3. Drive towards goal a few feet (selectable before start).
4. Turn control over to camera-based aiming algorithm.
5. Score ten balls with Sidewinder-like precision.

What has actually been tested:

1. Drive out a few feet.
2. Spin in place.
3. Ram the lockers at max speed.

We tested autonomous once before the robot went in the crate...

[pwilczynski]

We tested autonomous once before the robot went in the crate...

We didn't even get to test it. Everything is written, but.....you never know.

[Eldarion]

1. Drive out a few feet.
2. Spin in place.
3. Ram the lockers at max speed.
Nice! If you had been at the Wildstang Invitational, you could have seen such great auto modes as:
Back up full speed into the robot behind us
Run out towards our goal. Luckily it didn't shoot!
Position perfectly in front of the opponent's goal--and just sit there! :)

We have since corrected those problems, and can shoot all 10 into the center goal after plopping down in front of it.

[nehalita]

are you people using the camera for these?

[phrontist]

Wow, I'm shocked how many people can do somethign this year. Last year, had this poll been answered honestly, I'd say less then 50% could do anything in auto.

Of course there is a great deal of self-sampling bias, and chiefdelphi is not even close to representative of FIRST, but if this is even close to true it will be a much more interesting year this year.

I'm using the camera, and two quaderature encoders. I'm using the quaderature encoders in a somewhat... unusual way. The omniwheels are undriven, being used only for navigation (and as "casters"). Since omniwheels measure rotation only in one direction, polling the encoders gives vectors which can be integrated to give position. We'll see how it works ;)

[Eldarion]

are you people using the camera for these?
We are! :)

[Rickertsen2]

We can in theory:
*Drive closer to the center goal and shoot some balls in (not 10 though) using the camera. This is yet to have really been tested, but it should work once i tune our feedback loops a little better.
*Drive to some random place on the field using timing

[sanddrag]

I think our plan is to get about 10 feet away from the center goal, lock on with the camera, aim, and then fire about 7 balls in.

[Pavan Dave]

are you people using the camera for these?
Yep..

Thanks to the camera we will be able to shoot 10 balls in a very minimal amount of time.

-Pavan

[slickguy2007]

Our programmer worked hard to get the camera to work and it paid off. Not only are we accurate during the match, but during autonomous as well. We are able to drive out towards our goal and shoot all 10 balls in. We have tested it and it is working quite well. I dare to say we are ready for the regional next week? :)


GO 1403!!!

[Wayne Doenges]

Using the camera, we have 4 different scenarios that can be loaded, depending on our alliance partners.
Also with the help of our camera we can drive in from just about any angle and shoot balls into the center goal.
Once the bot is roughly lined up with the green light the driver hits a button and the camera and programming takes over and shoots.

[Team905Animator]

Dead reckoning, we have it set to shoot for the center goal from the starting area. During testing we had about 40-50% accuracy with it ;)

[Nuttyman54]

none written yet. we hope to use a combination of encoders and the YRG to drive to to the corner goals and dump. One thing i'm interested to see is if any robots will cross paths and screw up the aiming in autonomous...

[ForgottenSalad]

Shoot through the center goal from any starting spot, or dump balls in the correct corner goal closest to it's starting spot. No sesnors, so I have to use dead reckoning and timing and a little bit of luck to get to those corner goals...

Of course even though it's written, all of this hasn't been tested once yet, so it's all theoretical.

[nparikh]

I think the question is what can't we do in autonomous :cool: .

Thanks to the brilliant work of Bharat, we can shoot 10 balls in ~9.8 seconds, and yes using the camera.

for those of you who haven't seen it:
http://www.raiderrobotix.org/auto1.wmv

[Steve S.]

it cant do anything yet :(

but we plan on having it just coming out a bit and turning then shooting.

[dumbuya]

Good to know you have many Autonomous programs. I am team (1769) coach and we were unsuccessful doing the autonomus programming. If your team can assist so that we might try that during regionals.

We are rookie team in Kansas City, Kansas. We have a two wheel drive system. Those drive systems are the only thing we want to put in autonomous mode. Any ideas?

[team1852]

Oh man, we didn't even start programming Autonomous mode until about 2 days prior to ship date. :D

Right now the UNTESTED code is supposed to go forward and turn on our shooter. We're crossing our fingers that it'll works :rolleyes:

-Team 1852

[Eldarion]

I think the question is what can't we do in autonomous :cool: .

Thanks to the brilliant work of Bharat, we can shoot 10 balls in ~9.8 seconds, and yes using the camera.

for those of you who haven't seen it:
http://www.raiderrobotix.org/auto1.wmv
That looks quite familiar! ;)

On a side note, maybe proximity sensors (such as an ultrasonic ranger) will be very important to avoid an auto-mode collision?

[TubaMorg]

Our robot can hum quietly as it reflects on the disorder around it.

[Rickertsen2]

If i can piece everything together at competition:

*drive closer and shoot balls using the camera. (6 optamistically)
*drive in an arbitrary path programmed with a neato gui thing i made last year.

[lil_longo]

honestly i dont even know what it can do i hope it can do something by the time we get to trenton this week. i think we are trying to shoot in center goal but i leave that to the programmers

[lil_longo]

hey this post is just so i can get 50 posts soooo
YAYYYY 50 posts

[Nita]

it cant do anything yet :(

but we plan on having it just coming out a bit and turning then shooting.Sounds similar to what my team is doing. I think we have more than one plan, though.

[Eldarion]

I will be trying to add a "search and destroy" auto mode at the regional.
Luckily I had already made the sonar drivers before ship! ;)

[pyro20911d]

As of now, our bot can't do much in autonomous, but as far as I know, we're still planning on using the cmu cam at competition

[Dominicano0519]

We are extremely efficient at autonomus( 10 out of 10 balls)

[Tim Delles]

Wow, I'm shocked how many people can do somethign this year. Last year, had this poll been answered honestly, I'd say less then 50% could do anything in auto.

That is because this year the object that the camera is seeking and will not really change color depending on other lights. Also you aren't aiming for something on the floor, you are aiming above you, which also makes it a little easier than last year.

[Caleb A.]

Great job guys. I was able to very easily program to dump into the corner goals, and I tried to make it shoot into the goal. But since I only got to test the day before ship day and there were a lot of kinks we're probably just be doing the side goal. But I'm going to also try shooting over a short robot into the corner goal for fun.

[114Klutz]

Heh - please note this poll is hardly representative of robot capabilities. Around 10% of the bots I saw at Pac. NW can acutually shoot in autonomous.

As for our robot, we move forward, aim, control the turret, and then shoot balls through the hoop. We can normally get 3-4 balls into the goal.

[CraigHickman]

A second question to ask would be "what CAN it do in autonomous," instead of what do you PLAN to do. Many teams plan to do a large amount of things in Auton, but very few actually perfect their plans.

I can say from personal experience at our first regional (PNW), that it is VERY hard to shoot the center goal reliably during autonomous. It took us all of Thursday, Friday, and most of Saturday morning before we could get it to work. Only then were people recognizing this fact, and adding another level of dificulty: They wrote their auton to block us.

[114Klutz]

A second question to ask would be "what CAN it do in autonomous," instead of what do you PLAN to do. Many teams plan to do a large amount of things in Auton, but very few actually perfect their plans.

I can say from personal experience at our first regional (PNW), that it is VERY hard to shoot the center goal reliably during autonomous. It took us all of Thursday, Friday, and most of Saturday morning before we could get it to work. Only then were people recognizing this fact, and adding another level of dificulty: They wrote their auton to block us.

Indeed - the anti114 mode effectively prevented us from shooting.

However, as we have a turreting shooter - it wouldn't be hard to move the robot to the side a bit and unload from there.

[bobotics319]

Our team (319) BIG BAD BOB - can shoot 7-9 balls out of 10 into the center goal... in under 3 seconds. We use the camera to locate and lock on to the light in autonomous. It is not a theory; we were very successful at the BAE Regional in New Hampshire. Our team led the alliance to the finals where we brought home the finalist trophy. Also during the teleoperated periods, we used the camera to provide an auto aim feature that was also 70-90% effective every time. Maybe that is why we attracted so much attention from our opponents

[mwtidd]

our robot at first used dead reconing to get close to the target, and then lined it self up. We were shooting on average 8 balls per autonomous. The only problem is when you get involved with dead reconing defensive bots. the way we solved this problem was to add a wait timer, it worked twice and failed twice, unfortunatle the two times it failed were during the final round in manchester. i dont know if you have already competed but autonomous is i feel the most importent time in the game because often whoever won that won the game because it because very difficult to shoot when we were contested. good luck
-mike tidd
team 319