Autonomous Strategies

[Doug G]

While I believe we'll have a nice targeting shooter on our robot, I don't see the value in tracking down a trailer to score in it during auto, since it may be too close to a Payload Specialist. Say you score 5 of your 7 moon rocks, but the Payload Specialist puts in 5 or more into your trailer.

So anyhow all you tracking and scoring robots in autonomous, please come after us, our robot will be in the corner getting loaded (and will load up your trailer too!).

[Doug Leppard]

Originally Posted by Alan Anderson

I believe the most effective autonomous mode this year will be the one that results in no moon rocks being scored in the trailer your robot is pulling. There isn't anything the robot can do in the first fifteen seconds that can't be done later in the match, so scoring seven moon rocks quickly isn't imperative.

I would agree this is number one, as few as possible moonrocks in your trailer during auto mode.

BUT this might be, especially at higher levels, a prime time to put cells in your opponent.

But I am holdong off until I see how well our robot moves before I come up with a strategy.

[Tottanka]

We will be counting on those teams who have a lack of auto, or a fairly simple one that just drives away from the Human Players and try to track a stationary robot and get as close to it as possible.

Of course, that will only start after we spend 3-4 seconds running away from the payload specialist near us.

[Rick Wagner]

It may be impossible to get away from a good opposing PS shooter in autonomous mode quickly enough to avoid one or two moon rocks scored. Once away, however, the question remains, what to do fro the remaining 10 seconds or so. All three alliance partners can't go to the outpost to get an empty cell, but I think one should. The two others could lurk in their home corners until the driver controlled period starts. If a robot has good autonomous tracking and shooting, then it may be worth the risk to pursue a robot, even to its friendly outpost or fueling station because as long as it points in the right direction, a five foot tall robot will make it difficult for even a good human shooter to score.

[ExarKun666]

Originally Posted by Rick Wagner

It may be impossible to get away from a good opposing PS shooter in autonomous mode quickly enough to avoid one or two moon rocks scored. Once away, however, the question remains, what to do fro the remaining 10 seconds or so. All three alliance partners can't go to the outpost to get an empty cell, but I think one should. The two others could lurk in their home corners until the driver controlled period starts. If a robot has good autonomous tracking and shooting, then it may be worth the risk to pursue a robot, even to its friendly outpost or fueling station because as long as it points in the right direction, a five foot tall robot will make it difficult for even a good human shooter to score.

I like this suggestion. But some improvements might be 1 bot goes to empty cell, and the others could get into a good position for teleop, because I'm not so certain that your alliance corner is where you want to be at the beginning of teleop.

[Joshamuffin]

Another thing the robot could do, your robot could use the camera to aim at the trailer and head to ram it. This not only would throw other people's autonomous modes off, it would set us up to score with our basket. This is probably what our team will do.

[gorillamonky]

so far my team is going with the "RUN LIKE HECK" strategy during autonomous mode, although we may try to track down opposing robots

[rjmah]

Originally Posted by Abwehr

Totally agree. I expect that our team, and other veteran ones, will quickly go from team to team after receiving the match list in order to install at least a rudimentary autonomous mode.

Just to get everyone moving, for Labview teams using Basic Robot: Open the Autonomous block diagram and flip the while loop control from False to True. It will enable a 4 segment drive and turn program. All you need is to adjust speeds and also the motor outputs if you've changed them from default arcade. Hope to see every robot moving.

[waialua359]

Originally Posted by Alan Anderson

I believe the most effective autonomous mode this year will be the one that results in no moon rocks being scored in the trailer your robot is pulling. There isn't anything the robot can do in the first fifteen seconds that can't be done later in the match, so scoring seven moon rocks quickly isn't imperative.

The analysis sounds right on, however, we will still attempt to track, lock on and shoot during the auto period.
Since we are not looking to create a ball harvester and have human loading into our hopper, the more opportunities, the better. We would start our match during tele mode loading up.

[thefro526]

In my opinion Autonomous this year is going to go through some changes as the season goes on. Initially I think a large number of teams are just going to go with some kind of dodge or evasion strategy. I think this is going to be effective early on because a lot of teams are still going to be tuning their cameras and shooting systems if they plan on scoring in Autonomous.

As the season goes on and more teams become comfortable with the camera and associated autonomous modes we'll see a shift from defensive autonomous play to offensive autonomous play. This will be when a random dodge or zigzag autonomous may not be that effective because you could be moving within the range of one of the human or robot shooters.

As for my teams autonomous though, We've been talking about possibly 10 or more random autonomous paths so that you could never scout all of them, a program where we move in front of our HP and hope we don't get scored on by our robot, a program where we ram into the nearest target based on the starting position (Dead Reconing), and a program where we use our camera to track another robot and attempt to pin their trailer so that we can score if we want to at the beginning of Tele-OP.

[gorillamonky]

Originally Posted by Joshamuffin

Another thing the robot could do, your robot could use the camera to aim at the trailer and head to ram it. This not only would throw other people's autonomous modes off, it would set us up to score with our basket. This is probably what our team will do.

if you damage the trailer in doing so you will take a penalty (G30)

[gorillamonky]

has anybody given any though to the idea that getting anywhere specific during the auto period is going to be all but impossible?

my team put the new wheels on an old robot, and we found that precision steering only happens with human players towing the robot

[rjmah]

Traction Control is one of the big things this year. It's all in acceleration. Like driving in snow. If you just gun it, the tires spin. Nice easy acceleration will get you going faster. The challenge for autonumous is to get away as fast as possible without spinning the wheels.

[Travis Hoffman]

Where's the "drive to the center of the field and spin around in circles" strategy? Seems like that would be the most "compact" evasion strategy. Not sure you could get going fast enough to make it effective, though.

[kamocat]

I've found that it's actually quite effective at evading balls from human players, but it's also very touchy and prone to being disrupted by other robots. When you're jacknifed against your trailer, it's actually a lot easier to go backwards than forwards. (I suppose you could go around in circles backwards, but if you plan on switching directions much, you'll have to deal with the deadband in the jaguars)