recycle rush full strategy

[who716]

This is to discuss a full strategy, from auto to the last buzzer of the match, and everything in between! How is your team planning to play this game from start to finish?

[Rick Wagner]

We intend to have three stacked yellow totes in autonomous by having two robots carry containers while the first yellow tote is placed. Then the two other robots come back and get and stack their totes on the first using the vision system while the first robot carries its container. Finally all three robots end up in the auto zone for the full 40 points.

For teleoperated, naturally we will work with out alliance partners to make stacks of six with containers with litter on top. One robot takes a container to the human litter feeder, gets fed, puts it on a tote, then gets another while the other two robots make the stacks of totes. We will bottom stack. That is we will stack one with a container on it on one, then pick up that stack and put it on one, and repeat until we have a stack of six. Carefully we will put the six stack on the scoring platform. We think we can do this if we move slowly and carefully.

We will work with our alliance and opponents to get coopertition points during the qualification matches.

[ehochstein]

One piece of advice for any team following Robot in 3 days (Ri3D): realize that all teams come up with their own ideas for the game, but that doesn’t mean those ideas are the right answer. Ri3D doesn’t exist for teams to copy the robots; instead, it should help inspire teams with strategies or designs that could be used for the game. Make sure to question the ideas you see and try to find other approaches or possible improvements.

Before creating a robot, one must first decide what it must do; otherwise, things get messy—and `Snow Problem is no exception. To begin our design process, the first question we asked was what we wanted teams to get out of our Ri3D. One possibility was to show teams how to make a top-scoring robot that would likely be an alliance captain. However, we decided to take a less common route, as we felt that many other Ri3D were likely to build a primary robot. If it had already been done, we argued, teams wouldn’t learn anything new from our robot. We instead found a more specialized, simpler strategy that would be useful in the eliminations. With such an accessible design, more teams can build a design similar to ours and improve upon it with their own ideas.

To begin, we listed strategies we believed would reach the elimination matches. We discussed teams dedicating themselves to TOTE-stacking, robots filling the SCORING PLATFORMS with TOTES at one level, and robots placing the RECYCLING CONTAINERS on top of stacks. With all the ideas listed, we needed to know the effect of each strategy as well as whether it would carry an alliance into elimination matches.

We decided to go with the RECYCLING CONTAINER strategy in which we would grab, move with, and place the container on tote stacks. We felt this robot would help create a lot of points for an alliance. Each side starts with 3 containers, which could lead to 18 points if a robot put a piece of LITTER into each container. Using a HUMAN PLAYER STATION, we found that it was easy to manipulate the noodle into the container. In fact, we were confident that we could do so at the beginning of the match while our alliance stacked totes for more points. Once the totes were stacked, we could start placing RECYCLING CONTAINERS on top of them, essentially tripling the points from the stack of TOTES. The last part of this strategy was to use leftover time to get through the sea of TOTES to the STEP and grab one or two of the RECYCLING CONTAINERS on it for more points.

Once decided, we had to break down this strategy into a list of goals and priorities so we knew what to focus our time on. The list we came up with is as follows:

1). Must move
Sounds obvious but this means drive train needs to be solid

2). Grab and lift the RECYCLING CONTAINER and raise to tote level 4
After moving this is the most important part of the strategy. This does include getting the LITTER in the container since we’d move the container to the station and the human player does the rest.

3). Push TOTES
This is for pushing TOTES out of the sea of TOTES to get to the STEP more easily
The next few goals are things that are low priority but if left with enough time we want to achieve.

4). Pick up/stack TOTES
This can help our alliance with the coopertition points in qualifications or score TOTES if no one else can. It also means we can help score more TOTE points if the RECYCLING CONTAINERS are finished.

5). TOTE stack
The highest level at which to stack at created a lot of debate. We felt a minimum of four was a good place to start since we felt that this was very doable for TOTE stacking robots at eliminations. Trying to get the stack to 5 is something we didn’t think would involve a large jump from four and if we have an alliance partner who can stack to this height fast it would mean some extra points. One thing we kept in mind for deciding how high to go was whether it would be better to stack the RECYCLING CONTAINER on a couple tall TOTE stacks or many lower TOTE stacks. We felt that many at four is better than a few at five or six.

After finishing our list (and our strategy), we moved onto design, which is a blog for another time.

[nxtmonkeys]

My team is planning on using the kit chassis and building a sort of forklift for the totes and the cans, and also having an arm on one side that can "clean up" noodles/litter and deposit the noodles/litter in to the cans. I've already prototyped the arm, and it works surprisingly well.

We will also attempt to enlist a football player to my team as a noodle/litter-thrower.
We plan on using a couple of pool noodles as a bumper.
Unlimited mini-CIMs will help a lot.

This claw can easily grab noodles/litter and is relatively low-cost, but if you can get your hands on a 3D printer, you can print your own using the STEP file here. FYI, the link automatically downloads the file, so if you just want to look, use the other link.

Good luck everybody!

[tindleroot]

Originally Posted by Rick Wagner

We intend to have three stacked yellow totes in autonomous by having two robots carry containers while the first yellow tote is placed. Then the two other robots come back and get and stack their totes on the first using the vision system while the first robot carries its container. Finally all three robots end up in the auto zone for the full 40 points.

This autonomous seems VERY VERY VERY optimistic. Even getting one robot to do the things you mentioned is a feat few teams can accomplish, and to get three will be nearly impossible, save an unlikely Einstein-level alliance. At most I think we will see 24 points from 1 team that completed the Stacked Tote Set, plus all 3 robots sitting in the auto zone for the Robot Set.

[ZugzwangKG]

My team is going to stack the totes by lifting from the bottom of each stack. We're making a forklift, but rather than have it reach to the top of a stack to deposit a tote, it will only have to lift them up one or two totes high. This way we can have two stacks with totes following this model:
0...0
1...0
2...0
3...0
3...1
3...2
3...2+Bin
5+Bin...0
And just by grabbing from the bottom of each stack then putting it on the other, rather than lifting one at a time on top of a single stack.

[ishaves]

While still debating what startegy to follow, we did have the idea of prioritizing getting the bins off the step without having to push boxes out of the way. Since the handles are perpendicular it could be just a simple grab and pull then putting them aside or filling them with noodles from the player station. Just an idea.

[nixiebunny]

We decided that we'd make a robot that can stack and score both totes and trash cans. After all, having an alliance partner to depend on for part of the work in Week 2 is not a sure thing.

We also decided that having the robot process noodles is not high on our list of things to do.

Furthermore, the tote stack will be built on the robot, so that very little driving is required to make a tall stack. We'd like to be able to build stacks quickly from both the chute and the landfill.

We'd also like to do the autonomous TOTE STACK thing, although it may be tricky to get all three totes and deliver them in time. This requires good automagic tote grabbing ability, which we'd like to do anyways to ensure fast cycle time.

We may end up stacking the can upside-down on the totes to simplify the mechanism, so the noodle isn't an option in that case. We'll see.

[blturner]

We plan on a auto mode that gets and stacks all three totes all by ourselves. And another one that tries to get multiple containers. I don't see enough time to get three by ourselves. Hopefully the rookie teams can push a container and themselves into the auto zone. Preferably without knocking the container over as that will cost time later.
I think it makes more sense to let one bot work with the totes in their known starting spots rather than trying for autonomous coordination.

We may make the tote stacker program work both in front or behind the totes so as not to run into our partners.

After auto...it's stack, stack, stack. Time is points.

[Skyehawk]

I plan to put a lot of time and effort into autonomous this year. As people have said before me, time = points. While I have no doubt that teams will manage a three tote autonomous on their own this year you have to consider that a stack of six totes + container + noodle is 42 points. A perfect autonomous is only worth 32. Our team is thinking long term here, setting ourselves up for the rest of the game *hint, hint*. Anyway, good luck with your tote stacking endeavors.

[Edit] I ran the numbers, if everything goes close to how we plan we are looking at a total match score of 138 +- 10%.

[MrJohnston]

138 points? For just your robot or your entire alliance?

For an entire alliance, much, much more is possible. I'm curious how you would be adding that up for one robot. (3 x 42pt stacks = 126 - This would be very challenging, even with at 15 sec. start in autonomous).

[Skyehawk]

Originally Posted by MrJohnston

138 points? For just your robot or your entire alliance?

For an entire alliance, much, much more is possible. I'm curious how you would be adding that up for one robot. (3 x 42pt stacks = 126 - This would be very challenging, even with at 15 sec. start in autonomous).

Everything is perfect in that magical land called "Theory"

[bowmanb]

We think auto is going to evolve into a race for the center recycle bins. Who needs 24 pts when you can get seven noodle/bin/tote stacks. 7X42 = ??? That alliance will scoff at your Auto points.

[MrJohnston]

At the same time, I really don't think all three alliance robots are likely to be racing for the center bins - they'll get into one-another's way.... One should be staying back and collecting the autonomous points and preparing to stack the back bins and totes from the feeder station. Yes, go for the center recycling bins - but don't take that gamble and risk the back freebies.

[IronicDeadBird]

Auto: Automation of Preparation
This game is about resource management so 15 seconds while not a lot of time needs to be completely and utterly milked for every single bit of scoring potential you have. So depending upon robot design this could be doing bare minimum for moving totes or bins and focusing on positioning. It could be prepping certain mechanisms. I will be very interested in seeing what teams do with the 15 seconds they have because the freedom and protection the rules provide mean auto can be very detailed. There is 3 ball auto because of how resource management works. Balancing the time spent preparing, and serving is going to be key this year.

Tele Op: Small action Big impact
In past games you needed to be all over the field but the less space you take up the better. Yeah there might be a few wombo combo robots that work together but you get the most scoring by having the least amount of interference on your team. If you can score giant stacks but you take up a lot of space I feel like that will be less valued then scoring medium stacks and taking up less space.
Side Effects may include winning
While detrimental interactions within an alliance are bad (duh) since this game is about optimizing scoring a good alliance will be able to communicate needs that each robot has. The flip side to team mate only interaction is that while you don't want to mess up another robots flow, if someone asks you politely to pass the salt it is a lot quicker if you help out. (Once again without sacrificing scoring potential)

That being said I feel like strategy while completely correct in definition is not a word I will have in my vocabulary as much as I will have cycles and routines.
TLR
In resource management there is not ultimate idea there is only the understanding, mastery, and utilization of everything in the arsenal you bring to the table.