FRC 3128 Strategy CHARGED UP Takeaways

Competition Rules/Strategy
, ,
1

Hi everyone! We are the Strategy Department of Team 3128, the Aluminum Narwhals. We’re a team from San Diego, California, who won our first regional last year. We look forward to sharing our thoughts on Charged Up, and providing insight into Strategy’s role on our team during build season!

Offseason Review:

This off-season, we focused on improving our digital scouting systems, creating an in-depth curriculum, and teaching our new members the basics of FRC strategy. We went to two off-season regionals (Battle at the Border and Beach Blitz). During both regionals, we strived to test the improvements made to our scouting systems, as well as the new system we drafted for strategy at competitions. For more on the department, see our recently published whitepaper

Week 0/1 questions:

  • How feasible is attaining 5 links for the RP?
  • Are feeders valuable to an alliance?
  • Do we need a separate mechanism on our robot to reliably push game elements in low? (in case our manipulator breaks)
  • How heavy is traffic in the Thunderdome? (the upper middle area that lots of robots will have to pass through while cycling)
  • How impactful is defence? Are robots able to delay cycles reliably?
  • Is going around or through the charging station faster?
  • How many teams will score exclusively one type of game element?

Pitscouting points:

  • Drivetrain type, Robot Weight, Number of Motors, Motor Type, Vision:
    • These data points mainly provide technical information regarding each robot, allowing us to more accurately gauge a robot’s capabilities; for example, a robot with more motors is generally more apt for defence.
  • Auto, Auto Climb, Endgame Climb, Piece Type, Manipulator, Aluminum Assistance, Miscellaneous
    • These data points allow us to gain a better understanding of a robot’s capabilities, in addition to early insight into how we may be able to work alongside them in a match, making match strategizing more efficient. Additionally, it allows us to determine whether the team needs “Aluminum Assistance,” or technical help from some of our idle team members.

Scouting Data points:

  • Mobility, Auto High Cube, Auto Mid Cube, Auto Low Cube, Auto High Cone, Auto Mid Cone, Auto Low Cone, Auto Fumbled, Auto Climb, High Cube, Mid Cube, Low Cube, High Cone, Mid Cone, Low Cone, Fumbled, Climb, Defence Time, Penalty Count, Oof Time
    • We have compiled these data points for our scouting app, as we believe that they allow us to quantitatively and accurately rank robots based on ability.

What Strategy has been doing:

  • Immediately after the game was released, Strategy examined the game manual in greater detail. A way of doing this was creating a Kahoot for the rest of the team, testing their knowledge of R rules, gameplay, as well as general scoring rules.
  • Strategy created an everything list, aptly named for listing all actions that can be performed in a game, down to the ability to move. Using this, Strategy created a prioritized list of these items, determining which are more vital to success in Charged Up.
  • The department reviewed our priority list with our team’s technical departments, Mechanical and Controls, ultimately creating our finalized, and viable, priority list to guide the rest of Build Season.
  • Strategy also worked on an auto priority and everything list for Controls, to inform them of their work for the next six weeks.
  • We formed robot archetypes, which contained all robot types we could think of, from Dozer-Esque feeder bots to high-level robots. These helped us predict scoring averages for quals matches.
  • The department created a Driver’s Guide. This comprehensive manual includes strategies and cycles for particular scenarios, robot archetypes, ways of dealing with/performing defence, and general questions we have about gameplay in anticipation of regionals. Despite its name, a Driver’s Guide primarily serves to solidify Strategy’s understanding of the new game, though it can act as a catalog of strategies to implement during a match.
  • Strategy used previously created robot archetypes in order to make a paper game (a staged qualification match). Three fake alliances were created by picklisting our robot archetypes, and strategies against each were drafted.
  • To establish a range of scores for qualification matches, Strategy played a human game!(people on dollies to simulate robots on a field setup) With this simulation of Charged Up, we were able to estimate cycle times for various different robot types as well, which can help us create more viable strategies in anticipation of our first regional.

Predictions about gameplay

  • Climbing will be worth a lot of points and it is necessary for an RP. Auto will also be important as cycles will be much slower than in Rapid React, so getting pieces early during auto will be very important.
  • Defence may be influential since there is a large rectangular area in the upper half of the field which robots will have to cross through in order to go between the loading zone and the community.
  • There will be a lot of congestion in the cross-section robots cycle in
  • The manipulation of both game pieces allows for more versatility and higher efficiency when creating links.
  • Robot archetypes, such as a feeder/dozer-type bot, may initially seem strong because they are cheap, easy, and simple. However, feeder bots are very easy to defend due to them usually having to resort to pushing, which can easily be stolen. However, if a team is a low-row scorer, they could play a feeder bot and reduce a high-scoring team’s cycle time significantly.
  • Triple climbs may be widespread in higher-level regionals. To efficiently achieve this, at least 2 robots will need to have a swerve drivetrain.
  • Coopertition will also be very popular within this game because there are almost no downsides for both alliances. Not only do both teams have to go through fewer cycles to gain RP, but they also don’t lose any points for making links in this specified area.

Expect another update on this thread next Saturday, and we look forward to discussing everything and anything with you all :slight_smile:

20 Likes
2

Hello everyone! Staying true to our promised updates, here is the post for the second and third weeks of build season! Last time we focused on takeaways regarding CHARGED UP. This time, however, we focused more on game simulations; we also created a writeup for our CHARGED UP Scouting System, which you can see here.

To brainstorm what teams might look like at competitions, members of strategy played a paper game in which they designed 24 robots that they thought would appear during competition. They then made picklists ranking the robots, and we follow that with a simulation of alliance selection. Afterwards, members strategized how they would have their hypothetical team play during matches and debated who would’ve won - we went through an entire double bracket of elimination matches.

We also played a human game this week, giving us a better idea of how matches will play out in competitions. To simulate robots, members of our team rolled around using wooden dollies and were given certain restrictions on where they could place objects and which objects they could place. They used their hands to grab and place objects. The human game gave us a much better idea of how much of a role congestion will play during matches. Before playing, we thought that congestion would be pretty rough but overall not too bad unless defense was being played in the middle. However, while playing the human game, it was very evident that people were taking longer routes while cycling to avoid traffic in the middle of the field. It also gave us a better idea of just how big the protected areas of the field are. We think that it will be very important this year for drivers to be very aware of where protected zones are so they don’t rack up penalties.

Though it is not certain, we have been thinking of creating a scouting alliance with other teams attending the same regionals as us. We have already compiled a list of robots we would like to invite to our scouting alliance, but have not made any official decisions. This scouting alliance will mutually benefit allied teams by increasing data on a given robot and reducing the impact of inaccurate data. Additionally, allied teams will gain access to the scouting systems.
Since kickoff, we have worked on a Defence Manifesto for Charged Up! Generally, a Defence Manifesto is a compilation of predictions about defense in the game, as well as a collection of defence strategies to implement during a match. Complete with cycles, important rules to keep in mind, and instructions on how to deal and dole defense, it is analogous to a manual for drivers during matches, and allows Strategy to gain a better understanding of Charged Up!

For week three…

This week, Team 3128’s Strategy Department began working on a scouting alliance for Hueneme Port and San Diego Regional. Essentially, our scouting alliance hopes to invite other teams to use our scouting systems in return for a more reliable database, with ample scouting data for each robot ─ all teams involved will mutually benefit from it. This week, we invited two teams to our scouting alliance for San Diego Regional and one team for Port Hueneme Regional.
Strategy is also arranging scouting practice for our scouting alliance in preparation for our regionals. Meanwhile, our scouting system engineers are working to adapt our scouting system to the increased number of people using the system. Additional changes include a quick picklisting function, automatic verification, and auto-marking bad scouting data.
Another significant project Strategy is working on is a Strategy Workshop for local teams that want to learn more about strategizing at competitions. Our team will share how we collect data and make decisions at competitions. We also plan to upload a video of the workshop to YouTube. So far, Strategy has completed the curriculum for the workshop, planned out the logistics of it, and started advertising for it.

3128 Strategy Workshop
3128 Strategy Workshop720×900 204 KB

5 Likes
3

Here is a link to the form if you don’t want to scan the QR code.

1 Like
4

Hello! This is 3128 Strategy Department’s week 4 update! The best part of this week is that WE GOT CHAIRS FOR OUR SCOUTS!!! (We sincerely remind every team in the FRC Community to pay more attention to the health of your Strategy team, especially the scouts)!

This week, we created a “quality assurance” priority list. We found that the initial priority list, which we made at the beginning of Build Season, was too vague for our technical departments. In order to combat this, we deconstructed it and made each goal into multiple stages, labeled red, yellow, and green. If a subsystem is in the red, they need to be improved for competition. A yellow subsystem is satisfactory: if a subsystem is yellow, we might be at a disadvantage in a competition, but it would be satisfactory. Lastly, a green subsystem is ideal for competition. Using these checkpoints, we hope to help technical departments delegate their time more effectively in anticipation of our first regional at Port Hueneme!

Our Scouting Systems team has been enhancing our data analytics app, Neptune, by adding a picklisting function to the system. With this, we will be able to more efficiently strategize between competition days and in anticipation of alliance selection. Additionally, we have added more analytical tools to Neptune, including spider charts, performance trends, and five-number summaries for each robot. We have also been working on adapting our database to allow for data redundancy (i.e scouting alliances).

Additionally we developed a short curriculum for our Scouting Alliance in order to ensure that they understand how to use our scouting applications. A human game with our scouting alliance is scheduled for them to practice with the scouting systems, in lieu of actual matches to scout. We have also finalized our Strategy Workshop! We aim to help our local robotics community strengthen its understanding of strategizing at a competition. The workshop is taking place on Saturday, February 11th from 1-3 pm at Canyon Crest Academy. If you are not able to come, don’t worry! There will be a livestream of our presentation on our team’s YouTube channel and we plan to upload a video summary afterwards.

3 Likes
5
Hello everyone, this is Strategy’s Week 5 build season post!
We have had a few exciting updates concerning our scouting alliance: this week, we welcomed two more teams to our scouting alliance for Hueneme Port Regional! Last Saturday, we had our first meeting with our scouting alliance, we gave them an overview of our scouting system and practiced with the systems by doing a few human game simulations. We will also be meeting with them on February 18th and 25th to scout some Week -1/0 competitions (we’ll be watching these diligently, and next week’s post will be some takeaways from these early competitions).
We have also been fine-tuning and practicing our strategy workshop for Feb 11th (today!). Here is a link to the live stream if you would like to learn more about how our department functions and some tips on how to implement strategy during competitions. The stream begins at 1 pm PST.
2 Likes