Autonomous Strategy?

[EricH]

Originally Posted by Gerry Salinas

<G07> Center RACK position - After all six ROBOTS are in their starting positions and the DRIVERS, COACHES and HUMAN PLAYERS are in the ALLIANCE ZONE, the RACK will be moved (translated and/or rotated) to an arbitrary position so that the center of the RACK is within a 3 foot radius of the playing field center but the exact location and orientation of the RACK is unpredictable. After this point in time no ROBOT may be moved or repositioned until the match starts.

Nice try--I'll just look in Section 6.2.2: The Rack.

Originally Posted by 6.2.2

To constrain the motion of the RACK, the structure is connected by a set of loose chains to a small platform located at the center of the field. The platform is firmly attached to the floor of the Playing Field. The slack in the chains connecting the structure to the
platform permit the RACK to translate approximately one foot in
any direction, and freely rotate around the platform.

It's a 1 foot radius, 2 foot diameter.

[MattLi]

I would acually like to bring up another point. Some teams might have one or more programmers who can easily change their autonomous modes. For example. Team 245 (us) is planning to have 4 autonomous modes.

1. Score a keeper
2. Block another robot from scoring a keeper
3. Drop the keeper and set ourselves up to begin scoring ringers
4. Go to the opposing team's home zone and knock over all of their keepers

I would like to bring everyone's attention to auto mode no. 4. Many teams are going to be planning their strategy around the fact that the ringers are going to be upright. If we knock them down they can't pick up the ringers that are on the field and have to resort to the ringers fed through the chute.

[EricH]

Originally Posted by MattLi

4. Go to the opposing team's home zone and knock over all of their keepers

I would like to bring everyone's attention to auto mode no. 4. Many teams are going to be planning their strategy around the fact that the ringers are going to be upright. If we knock them down they can't pick up the ringers that are on the field and have to resort to the ringers fed through the chute.

I think that this will be a lot less successful than you think.

[MattLi]

Originally Posted by Abwehr

The more I do the math, the less autonomous mode seems to matter.

I made a scoring program that computer spoiler value (http://www.chiefdelphi.com/media/papers/1918) and if you play with it for awhile, you'll see that only VERY rarely does a single location have a significantly higher scoring value than any other on the rack. Usually, there are 3-4 other locations that result in the same spoiled point value...you'd just go for one of those instead.

The more I play with your program the more I agree with you. I doesn't really matter where you place the spoiler in a row. It seems that any place you put the spoiler takes an equal amount of points away from the opposing team. The keeper now seems to have less value then we would have thought.


And yes knocking over the keepers may be ineffective however, when faced with a team which can only pick up ringers from an upright posistion this strategy may yet be effective.

[GVDrummer]

One of the modes we are using is to drive around the rack and sit in from of it to block the other teams fom scoring. But we can also you the camera if we want. We are working right now on using two at the same time to form a triangle to the point of the target.

[Doug Leppard]

Originally Posted by Gerry Salinas

<G07> Center RACK position - After all six ROBOTS are in their starting positions and the DRIVERS, COACHES and HUMAN PLAYERS are in the ALLIANCE ZONE, the RACK will be moved (translated and/or rotated) to an arbitrary position so that the center of the RACK is within a 3 foot radius of the playing field center but the exact location and orientation of the RACK is unpredictable. After this point in time no ROBOT may be moved or repositioned until the match starts.

So, yeah, that camera is kind of important.

I agree that is what the rules say. But in a drawing is shows 6 inches of chain to allow play and in the rules 6.2.2 it says "The slack in the chains
connecting the structure to the platform permit the RACK to translate approximately one foot in any direction, and freely rotate around the platform."

So is it 6 inches, or 1 ft or three feet. Seems the rules and drawings say all three.

Not that it matters, it is tough either way.

[T3_1565]

Originally Posted by MattLi

I would acually like to bring up another point. Some teams might have one or more programmers who can easily change their autonomous modes.

This is our plan as well we are going to have an autonomous mode depending on the auto modes of our team, this way we don't get in the way of any other robot that can score (unless of course we want to be in the way )

[DRH2o]

Scoring is our plan. However the best laid plans of mouse and men ... You know what I mean

[Donut]

Originally Posted by GVDrummer

One of the modes we are using is to drive around the rack and sit in from of it to block the other teams fom scoring. But we can also you the camera if we want. We are working right now on using two at the same time to form a triangle to the point of the target.

Alot of teams are saying they will do this, however they make it sound like this is unbeatable. Do all the defensive auto mode teams really think they're going to be able to sprint almost twice as far as their opponents (35 feet compared to 20 feet), turn, then drive again to hit them, before one of those teams gets a keeper on? I can see it being effective sometimes, but not every single match.

On a side note, we want to have 12 different auto modes minimum this year.

[Lil' Lavery]

If all goes as planned, 116 will have a plethora of options to chose from when starting autonomous, basically boiling down to 5 or 6 basic functions.

[BBnum3]

We hope to have the ability to put on a keeper as well as defend by going to the other side of the rack. I do think that autonomous this year is not going to be as important, or as doable. Last year a robot that could consistently dump 10 in the lower goal in autonomous won autonomous most of the time, and that ten point bonus was huge.

I think the bonus gained by scoring a keeper won't be that great. It seems to be much more difficult to put on a keeper than score in the lower goal. The camera needs to be utilized, and there's the possibility of the rack being wiggled. That said, my team will still be trying for a good autonomous because it is a good challenge.

[akshar]

Originally Posted by MattLi

4. Go to the opposing team's home zone and knock over all of their keepers

I would like to bring everyone's attention to auto mode no. 4. Many teams are going to be planning their strategy around the fact that the ringers are going to be upright. If we knock them down they can't pick up the ringers that are on the field and have to resort to the ringers fed through the chute.

assuming you mean their ringers....beacuse that is what is up against the opposite alliances' home zone, you would be knocking down your own ringers and not theirs.
not all teams are designing their robot so that the tubes have to be upright in order to be lifted.

[Daniel_LaFleur]

Originally Posted by Donut

Alot of teams are saying they will do this, however they make it sound like this is unbeatable. Do all the defensive auto mode teams really think they're going to be able to sprint almost twice as far as their opponents (35 feet compared to 20 feet), turn, then drive again to hit them, before one of those teams gets a keeper on? I can see it being effective sometimes, but not every single match.

On a side note, we want to have 12 different auto modes minimum this year.

Sprinting will not be needed. Most robots trying to put on a keeper will be computing distance and angle of the illuminated target. By homing in on said target, they will not be moving at quick speeds.

My guess is that any scoring done in autonomous mode will happen between the 10 and 15 second mark, plenty of time to get a defensive tank (ave 5' per second) over there and bump the offensive robot off their mark. Remember, the offensive robot needs accuracy, not the defensive robot.

I can see it being very effective, but not effective every single time.