#4744's Alternative game strategies for Rapid React

Hey!

My name is Guy, and I’m a mentor at #frc4744 Ninjas from Israel. I’m writing on behalf of the team. We’ve decided to publicly reveal a part of a strategy method we used this season. We hope that by revealing the method we will help more non-top-tier teams like us to create better-than-previous robots and use their already existing robots more effectively, as happened to us this season. Although it was a post-covid season and with a team of only 12 students, the team has managed to create one of the best robots in the history of the team.

This post will be divided into two parts:

  1. Reveal of the essence of the method. We offer teams that are not competing anymore this season a long-term solution for the seasons to come. This part is summarized, but we will happily help teams who will reach us to make use in the method.

  2. Reveal of alternative strategies for this season that have been developed by the method. Because we are not competing anymore this season, we want to offer to still-competing teams other ways to play differently, to make their already made robot more effective on the field, and by that to increase their chances to succeed in this season’s championship.

  3. Essence of the method:

A month before the Kickoff we conducted a meeting revolving around the strategy making method. The method, proved effective, can be summarized by the followings:

  • Every team has a level, driven by a lot of factors (i.e. amount of students, previous experience, training, etc.). a team can “level up” mainly in the off-season period and between competitions, but it’s very hard to “level up” in the robot building period. The chosen strategy should fit not only the manual, but also the level of the team, or they might find themselves with a robot that is not functioning as desired in the good case. The strategy shall create a smart allocation of the team resources.

  • The Quals and the Playoffs have different goals. The goal of the Playoff matches is to score more points than the opponent. The goal of the Qual matches is to get as many ranking points as possible, either by completing special missions or by winning the match. Receiving more points than the opponent in the Quals is meaningful, but nor as crucial as in the Playoffs. Another goal of the Quals is to show the robot’s and the drive team’s abilities.

  • The majority of the points in the match are scored by using a game element that can be thought of like an economic resource, and time in a match as the currency. When the amount and accessibility of the element is great, and less robots compete on them, the price of scoring points is cheap. When the amount and accessibility of the element is poor, and more robots compete on them, the price of scoring points is expensive. different things in the game (i.e. rules, field elements, robots) may change the amount ,accessibility and the competition on the game elements and thus the price of scoring. Places of scoring (i.e. safe zones for shooting) can be thought of like a resource too. This is why good offencive robots can score from almost everywhere on the field.

  • Most of the robots will play offense or defense. Some offencive robots will try to score all-around the field, a little bit from everything. Some will specialize in a certain mass-scoring or rp making mission, סn other mission’s account. The defencive robots will try mainly to control the “locations” resources by preventing, hindering or pushing other robots off the locations. From time to time you can see a robot assisting another to climb, but it depends on the game and rules.

  • This season, as the match plays, balls tend to accumulate on the farther side of the field (this can be proven statistically).

Behind those points stands a deep method that helps, among others, to determine the level of the team and to analyze what can help or prevent us from controlling the game resources. We will be happy to help teams that will reach us to get to the depth of the method.

  1. Alternative strategy to this season

The method, previously explained, helped the team to think on some alternative strategies that rarely seen in play this year:

  • A far-operating robot: because balls get farther away from the alliance, a robot that can pick-up and score balls from the other side of the field can make an effective scoring method.
  • Delivety-bot: as previously claimed, the game element can be thought of as a resource. a robot with a good intake and outake but not very good shooting abilities can help its alliance by making the balls more accessible to its alliance member, rather than trying to score it itself. In advance, there are optic dead-zones for every close-by team station in the alliance wall. The third station has a complementary field of view for those exact dead-zones, meaning a 2nd pick robot can make the balls accessible for the more offensive alliance captain and 1st pick. This strategy can’t win a match on its own, but can help even a medium-shooter alliance member to score more points.
  • Snatcher-Hider: This strategy is the exact opposite to its prior. If the goal of a delivery-bot is to make balls more accessible to its alliance members, this strategy’s goal is to make balls less accessible to its opponents. The snatcher part makes it harder by increasing the demand for the same type of balls as the opposite alliance (therefore increasing the price of point scoring), and the hider part makes it harder by decreasing the supply of those balls by hiding them. Hiding spots can be found in the exact same places of the other alliance optical dead-zones: on the robot itself, near the truss legs, near the hub fenders and out of the field via the terminal. Once again, this strategy can’t win a match on its own, but can help in creating a point difference between the alliances. Here is a video of us testing out this strategy, you can notice that from the moment of placing, those balls haven’t got picked up by the opponents.

Drawings of the optical field of view and dead-zones can be found at the end of the post.

On this occasion I would like to call for all of the non-top-tier teams out there: it is the time for us to play smarter. It is the time for us to return to search the gaps in the manual, to explore the “may be done” and not only the given “can be done”, to make the alliance picking more complicated than a preparation of two lists: offencive and defencive robots. It is the time for us to make the games and matches more interesting to see, play, and plan tactically. It is the time for us to challenge the top-tier teams to get better themselves.

We will be happy to assist teams that will reach us about the strategies, or to deepen their understanding on the method and process.

Also, full drawings of the optical fields of view and dead-zones are available down below for anyone to use, just keep the team’s logo for credit for the work.

Good luck for all teams, both the still competing teams and the next season competing teams :slight_smile:

**Optical analysis of fields of view and dead-zones, created by #frc4744 **

24 Likes

Very well thought out framing for strategy! Its really useful to see your bot placing the opposing balls into the dead zones and having them stay dead!

2 Likes

Thanks!
In my opinion, this video really shows the effectiveness and potential of the strategy.
Honestly, the opposing alliance had really good offensive robots, and our scouters estimated their ball count to be 22-25, thus reaching the ball amount needed for the RP.
With this strategy their ball count was only 12, without us playing regular defense that is susceptible to fouls and damage for the robots.
It may not be enough to stop the high-tier robots from getting the RP, but may come in handy in the Playoffs, as a way to slow the opponent alliance and make the crucial point difference.

2 Likes

Interesting. I believe some teams have received a penalty under G404 (blue box part b) because there were four opposing cargo in their hangar when they started climbing before 30s. See G404 and Ball Entropy @ CVR QF2-1.

1 Like

Absolutely right!
This is one of the deeper insights my team reached.
The alliance may climb only in the last 30 sec of the match, or make sure their climb isn’t restricting more than 3 balls under G404.
By the way, there is no restriction for climbing with 2 opposing balls on the robot, taking them completely out of the game.
Thanks for the heads-up for any team who might want to implement it!

Balls right against the driver’s station wall are also invisible…
A really skilled human player can indicate where balls are, assuming anyone is actually paying attention :wink:

4 Likes

You are right.
We didn’t add this to the list (and also the dead-zones created by the opposite hangar) because it may decrease the cycle time for the opponent alliance, rather than increase it, having the opposite effect.
Your alliance wall is also quite of a dead-zone (mostly from the middle station to the left towards the hanger).

The human players, cameras and auto-ball-targeting make it harder for the strategy, but from our drive team experience - even when you have cameras and a human player who can signal, it is still hard to drive the robot correctly.