I’d like to take some time to discuss our game strategy and thought process behind the design.
This game, Aerial Assist, is truly unlike anything FIRST has ever done. The game is simple, intuitive, and at first glance, there aren’t a lot of specializations – robots need to get a ball, then get rid of it.
We looked at specific actions, and found seven categories: Floor load, Ground pass, Eject over truss, High goal score, Catch, Low goal score, Goaltend. From these, we ranked them in order of importance to our team. We figured the main priorities for our design are 1. High goal score, 2. Floor load, 3. Ground pass, 4. Eject over truss, 5. Catch, 6. Low goal score, 7. Goaltend. Here’s our reasoning:
1. High goal score: We can start autonomous mode with a ball. The very first thing we need to do with it is score it in some way. Before we can accept a ball, we’ve got to get rid of the one we’ve got in a beneficial way. Also, all the assisting in the world doesn’t matter if the ball isn’t scored at the end. For these reasons, we ranked this highest.
2. Floor load: We figured that if a ball isn’t in a robot, it’s most likely on the floor. We didn’t want to restrict ourselves to need a passing alliance robot or human player skill to be able to intake a ball. We also thought if we had a floor loading system, we could probably reverse it to ground pass.
3. Ground pass: We believe most robots will be designed to take balls off the floor – an assumption that was supported by all the 72-hour robots. If we can transfer the ball to our alliance partner in a controlled, quick fashion, that would be best.
4. Eject over truss: This is a big 10-point move (20 points if an alliance partner can catch), so it ranks highly. But it got knocked down because we figured if a mechanism can score into the high goal, it can easily be adapted to loft a ball over the truss.
5. Catch: Certainly a critical ability that any top-tier alliance will need, but it is highly specialized and, to do it well, we believe would require a lot of concessions in overall design. We thought that with our resources, to build a specialized catching robot, it would severely limit other functionality purely based on robot real estate. If our robot ended up in a position where it could catch, that would be great, but it doesn’t really fit in to the gameplay strategy of the above abilities. If we are going to design to catch in this game, we need to be really great at it, and we figured that would necessitate a swerve or octocanum drive train. We’re not quite in the 16/488 echelon yet.
6. Low Goal Score: If we can ground pass, we can probably score in the low goal. This isn’t a primary objective for us, especially since it’s only one point.
7. Goaltending: We don’t believe a goaltending robot will have a place on top alliances; maybe in the 5-8 seed range, with alliance captains who are thinking they need to shut down the powerhouse. For this game, we’d rather win a track meet than a defensive standoff. Also, we thought that given the size and velocity of the balls, it would be very difficult to design and build a working goaltending system with our resources. Finally, there are two areas of the field that are pretty much undefendable – the high goal directly above the low goals.
We’ll blog again soon.