Autonomous Strategy

[jojoguy10]

I'm not a huge fan of discussing strategy on CD because I think it ruins the surprise (since we all know what everyone is doing), but I think this is worth discussing.

I wanted to know if any teams were planning on working with their alliance partners during the autonomous mode. To be more specific, here's a scenario I've been thinking about.

ROBOT 1 picks a yellow tote and puts it on the LANDMARK. ROBOT2 (maybe after 5 seconds) picks their tote and tracks the first tote to stack on top (creating a stack of two yellow totes on the LANDMARK). Finally, ROBOT 3 picks their tote and tracks the stack of yellow totes and stacks them on top. The 15 second autonomous ends.

I'm curious how many other teams have thought about this plan and what they plan to do with it (or just totally scrap it?)

[Ginger Power]

I'm thinking this version of autonomous stacking will be incredibly rare. It requires all three teams to have vision tracking, and usually if teams are advanced enough to have vision tracking they will have the capability to do auton on their own. A much more common auton strategy, as has been discussed on other threads, will be teams going for the Recycling Containers from the step IMO. I think that if the stacked tote set is accomplisjed, it will be accomplished by a single robot.

[Munchskull]

Originally Posted by Ginger Power

I'm thinking this version of autonomous stacking will be incredibly rare. It requires all three teams to have vision tracking, and usually if teams are advanced enough to have vision tracking they will have the capability to do auton on their own. A much more common auton strategy, as has been discussed on other threads, will be teams going for the Recycling Containers from the step IMO. I think that if the stacked tote set is accomplisjed, it will be accomplished by a single robot.

I agree. The three tote auto will most likely be done by one fast robot. The amount of synergy that would be required to have three robots do the three tote auto is unprecedented. I predict that it will happen once this whole season.

[Jared Russell]

Originally Posted by Munchskull

The amount of synergy that would be required to have three robots do the three tote auto is unprecedented. I predict that it will happen once this whole season.

I think you're being overly optimistic.

[JesseK]

We have a design that is easily capable of a 3 tote auto on paper and in CAD, but it definitely requires partners to play along no matter how well we build it.

[pafwl]

I agree with the original premise. Middle robot pushes to landmark. Somewhat easy.
Right robot goes to landmark and drops at 8-10 seconds. Back ups.
Left robots goes to landmark by 12 seconds and drops at 13 seconds.

We will try to have all of these woking and will be available to anyone who wants to play.

Jared I assume you meant one robot doing it is overly optimistic.
I have learned, if Jared is telling you it is overly optimistic take that to the bank.

[BigJ]

I think he means 3 robots cooperating to build an auto stack without something going wrong for a relatively high-precision group maneuver is overly optimistic.

Edit: especially when there are much more useful things for teams to be doing if they are of the caliber that could do the group maneuver without something going wrong

[Amar Shah]

I could see two robots doing this, with only one needing to use computer vision. The first robot only has to move one tote to the auto zone, while the second robot picks up two totes and uses computer vision to find the second tote and carry the stack to it.

[Munchskull]

Originally Posted by Jared Russell

I think you're being overly optimistic.

True

[troxbots]

I could see a strategy with a stack if two robots can push the first two cans out of the way, one robot drives across stacking up all bins, and pushing last can in to auto zone.

[MrJohnston]

I don't believe it's worth the effort to design a three-robot-tote-stacking autonomous strategy. First, you'd have to have two partners who have designed the same thing - and each chosen different complementary roles within that specific routine. The odds of such robots every being grouped with you are very, very low. Secondly, even with such a pairing, you'd have to practice setting it up and work together to adjust the routines so that the efforts are perfectly coordinated. Third, if you have three robots with that level of coordination, there are more useful things to be doing during autonomous and using all their efforts for 20 points is likely not worth the effort.

I also believe that the easiest way to do a three-tote autonomous routine is for a single robot to do it completely on its own - moving the recycling containers out of the way, picking up all three totes, etc. Adding a second (or third) robot to a small area increases the chances of a little bit of mistiming, collision, etc.

[Munchskull]

Your alliances time would be best spent having one team pull a solo three tote, then the other two working for the center containers and moving them in to the auto zone with the ultimate goal being to get at least a container set all from the step, a robot set and a three tote, totaling in 32 points.

[troxbots]

If your alliance partners can just move forward immediately, pushing the cans, and if you design your strategy, and path of movement correctly you can get all of the points as Munchskull described.

[MrJohnston]

Originally Posted by troxbots

If your alliance partners can just move forward immediately, pushing the cans, and if you design your strategy, and path of movement correctly you can get all of the points as Munchskull described.

"IF" your alliance partner can - and without disrupting the totes. I believe it would be easier for a single robot to grab all the totes and recycling containers, scoring 28 all by itself.....

[troxbots]

I like to hope they can at least move forward, but yes you could. I would be impressed with a robot that can move all three totes, and all three bins, but its possible