Day 1, 2

The most important days of build season have now come and gone. I hope they were used wisely!

This past Saturday, the new game, FIRST Destination: Deep Space, was revealed to teams across the globe. Saturday and Sunday are all about learning the new game, the rules, the points, and building your teams baseline strategy. On these days, we analyze what we want the robot to do and what tasks we will complete to score points. We think about point cycles, point ceilings, qualifying points, ranking points, district points, and more. The important part of these discussions are answering the 'what' question, not the 'how' question.



This year, as in past years, we invited some friends over to play! Kickoff weekend, we had our friends 4681 Murphy's Law and 5803 Apex Robotics join us for our rules, points, strategy discussions, and our gameplay demos. We are always glad to have friends over, and collaborate on ideas, understand the rules and gameplay, and share in some good food!
Our story begins bright an early, 6:30am, with a cup of coffee and the reveal of Destination: Deep Space.

Rules and Points

When a new game is revealed, a lot of information is thrown at everyone, students, mentors, parents, volunteers, etc. We then throw all this information at our co-workers, family members, and random people at the grocery store. The new game, the theme, the game animation, the game rules, robot rules, competition rules, are opened up to everyone for analysis and breakdown. One critical lesson that FIRST teaches is how to take a gigantic, huge, seemingly complex task, and break it down into small, bite-sized, manageable chunks.

4911 starts this process almost in silence. Each team member gets a copy of the rules on a phone, laptop, school computer, or smoke signals directly from FIRST HQ in Manchester. We take enough time for each student to read through the rules, focusing on three main sections: point scoring (and deduction), gameplay and penalties, and a look through common rules, to see minor changes for this year (robot size, competition points and tie-breakers, etc).

Once we understand the rules, we gather back as one. We draw and describe the field, every method of scoring points, throwing in some quizzes and questions about rules, penalties, zones in the field, human player tasks, and more. This ensures that all students understand the core of the game, and start building creativity and inspiration for our strategy discussion. We compare options for scoring in autonomous and endgame, starting to list out all possible tasks that a robot can perform. These are written as very specifically-generic statements: collect the cargo from the platform, score a hatch cover at low level, cross the center line during sandstorm. This is done very intentionally to answer the 'what' question, without getting caught up in 'how'. We do this so that we are able to think about all possibilities, without getting caught up in mechanism design and feasibility. This comes with a caveat that some ideas can be too far out there.... deep in space... and do not promote positive discussion or additional ideas. Teams will find this boundary on their own.
Our rules and points discussion ends with a chart of all possible ways to score points, a maximum theoretical score, and a list of all possible tasks that a robot could perform.

Strategy Thoughts

What do we do with lists? We make more lists!
Taking the list of points and tasks, we can start looking for point cycles - repeated tasks to obtain game pieces, manipulate them, score them, and return to a loading zone. Point events are one-time scenarios to generate points, such as crossing the center line during autonomous, or moving to a platform in endgame. There are also single point events that can be maximized, usually in autonomous or endgame. For Deep Space, we looked at crossing the center line during the sandstorm, and ending the match parked on the platforms.

Each of these cycles will be compared for points, and given thoughts about realistic times for each step in the cycle. Using this, we can estimate how many points a robot playing each cycle could potentially score during a match. This helps drive our decision-making process. Cycle A can score more points than cycle B, so we should start thinking more about the tasks required in cycle A. We build a spreadsheet with variables for cycle times, hatch scores, cargo scores and sandstorm/endgame points. These spreadsheets allow us to quickly compare different cycle options, benchmark scores, and verify wants/needs from our robot requirements. This leads to... another list! The tasks required to complete cycle A become ranked, and create our robot priority list. This will be the living list that drives our entire engineering effort for the duration of the season.

We prioritize these tasks for 2 reasons: The order of the list will dictate engineering effort and time, as well as drive practice effort and time. The order of this list will also help choose what we consider a 'minimum viable' competition robot, or inform us of tasks/mechanisms to remove if build season passes us by too quickly. This list is a living list, as priorities may change as team updates occur, Ri3D are revealed, MCCC are played, and Week 0 and early competitions are played.

Predictions and Benchmarks

We will never be fast enough. The chassis will always be too slow, intakes will always be too slow, scoring will always be too slow. This is ok!
Our goal is to think and brainstorm - realistically - as to how competitive teams in our area, and globally, will strategize. How fast will their cycles be. How many objectives will they score in Sandstorm? (Will they run autonomous during the Sandstorm?) These benchmarks generally come from looking at past games. We predicted autonomous cubes last year, how close did our prediction come to reality? These predictions will become cycle time benchmarks that we test against our prototypes and drive practice.

Along with thinking about specific strategies we feel other teams may delve into, we also want to think and plan how the game will evolve over the course of competition. Week 1 district meets will play very differently from week 8 championships. Qualifications will play differently than eliminations. Understanding how the game evolves will also help focus our mechanical/software prioritization and driver practice as we progress through the season.
We will always want to be faster...

Practice Play

Day 2 started with a recap and re-understanding of our entire discussion from day 1. Having a night to sleep on everything and let it sink in helps ensure we all understand the decisions made for strategy, and robot requirements.

To continue to understand how the game will develop and play, our next step is to, well, play the game. Using human stand-ins for each robot, we defined several common robot capabilities, and played Deep Space. One, it was a lot of fun. Two, it helped further our understanding of our point cycle estimates, defensive play, helped us start practicing overall alliance strategies utilizing different robots, and more. We ended the day recapping the lessons learned by our gameplay testing, and assigned students to engineering sub-groups. The mantra, steal from the best and invent the rest, is really hitting home with the team this year. Last year we read about another team that sat in a big circle and pow-wow-ed. We stole that idea. I'm pretty sure that idea is going to get us back to Einstein.

This weekend saw the introduction of the new game, the breakdown of the point scoring (and deduction) opportunities, a list of all/most robot tasks, our prioritized robot task list, a breakdown of the robot requirements into sub-system requirements, student team assignments, and some fun human-played gameplay. Only 44 days to go! Til bag-and-tag! Because, you know, there's only six weeks in this program....

We don't usually do this, but have a quick teaser: