With team 4152 one of our goals is to improve our strategic design. Fundamentally we want to analyze the game effectively and focus on what we want to do. For Infinite Recharge we debated a specific aspect of the game. The ability to go under the control panel resulted in some teams choosing to build short robots. Other teams (we fall into this category) chose to build a taller robot (easier to climb, better shooting angle).
I feel good about our choice - we competed at Georgian College and made the top 8 and won the industrial design award. Our design choice was primarily driven by the shooting angle a taller robot provided. Our tests indicated that we would consistently be able to shoot in the outer target and have a better consistency at the inner target. Keeping tall allowed us to simplify our storage system and climber.
I was interested in what the data said so I analyzed the 53 events that had ranking information from this past season. Out of those events 29 of the number one ranked teams had tall robots (55%) while 24 robots (45%) were short.
My questions for the Chief Delphi community are:
What were the factors that you used to determine whether to build a short or tall robot?
Do you feel you made the right decision - if not, why would you change your mind?
We were unlikely to have the technical skills to package everything we wanted to do inside a short bot.
We felt confident a robust drivetrain could handle bumps in the center with ease, and not add large amounts of time to our cycles
We took the gamble that there would be at least a few other short bots who could use the under-control-panel lane, so as to not clog the center.
The shooting angle was better, especially since we wanted to take the long shot.
The biggest thing we ran into was our primary long-shot position ended up being right behind the control panel, effectively blocking that short-bot path. We had to get good at getting in and getting out, and doing our best to stay out of the way of alliance partners.
I do feel it was the right decision for our team, for this season. But not necessarily for everyone… If you can package all your tall-bot abilities into a short bot, you’ll have an inherently better machine (one additional downfield path it can take). But that’s a big IF.
The packaging aspect was crucial for us. We knew we could package for a tall robot effectively. The small robot added aspects that would have taken away from our practice to ensure a consistent shooting mechanism.
I think the first thing you should do is ask some basic strategy questions,
Will not being able to go under the trench severely hurt cycle times?
If defense is being played will we suffer not being able to go through the trench?
Where will we shoot from that can avoid heavy defense (i.e. trench run) and without going through the trench will we lose time getting there
Once those are established you can start thinking about packaging, because frankly if you can’t produce a short robot that will shoot somewhat as consistently as a tall robot and have no frequent jams, is there really a point? My thoughts going into a second season of Infinite Recharge is being a newer team with less than 15 kids, becoming a short robot would be tremendously hard to be consistent. Our solution for this is probably going swerve so that cycle times won’t hurt as much with the very little practice we get.
I totally agree with this, and can’t wait to see if champs happen to see robots try to get those two balls before other short robots. Personally I think it will turn into a just let the opposing alliance get them as they can fit 5 (having 3 already) and you don’t want to risk ramping over balls to protect them, possibly getting pin points from it.
We decided to go short for better cycle times. We shot from the target zone so getting our shots blocked was not a worry for us and it worked as a strategy making us one of the top 3/4 best shooters at Palmetto.
Our only problem was the lack of a climb which I doubt we would have solved before Palmetto if we did go tall since we prioritized the shooter first. I think that if everything didn’t go to hell in a hand basket we would have gotten a climber on and could have ranked very well at Smoky Mountains.
First of all, kudos for taking the time to understand how and why decisions are made. I think this analysis is one of the most effective ways to improve competitively.
For our team, we ultimately built a short bot, but traded high shooting for this ability. With the resources available to us, we felt that we would be unable to build both an effective climber and and effective high-shooter, and decided that the climber was more critical to success in this game (both because of the quick points it afforded and the possibility of additional RP). We felt we could still be very competitive with a low goal scoring machine playing a supporting role by helping to start the positive feedback cycle by rapidly scoring a large number of game objects (which we also felt was the easiest way to the second bonus RP).
Overall, going short may not have been the correct call for our team, and we even were heavily considering making a modification to our climb system at our event as the result of climber issues that were caused by the short configuration. We ended up (mostly) solving the issues we were seeing, but there would’ve been more discussion about reliability had the season continued.
For us, CD helped convince us. Our team was pretty set on a mecanum robot for this game. As we were deliberating and weighing the pros and cons, I brought up all the points from CD. There is a hearty anti-mecanum feel to many polls here, but the discussions gave some compelling reasons to at least be certain why you chose them if not avoid them altogether. We lisyened and weighed everything and determined mecanum was best but… We did want to protect against the biggest issue (defence). So we determined either we needed to make a butterfly drive or stay in protected zones. A small bot design seemed easier for us. However, we did not get a chance to compete and are needing to re-engineer our hopper (it is smaller and not as efficient as we would like). I do not know if we will get the chance to compete with it, but the process was incredible and we will be a better team for it.
One other thing to consider: Defendability of shot. In general, to ensure you can shoot over a defender, you need your release point to be as high up as possible, and as far from the goal-facing bumper as possible. Though we didn’t have enough matches for it to really play out, there were a number of robots we saw that could be defended against by just stopping between them and the goal. Forget bumping or nudging, you’d block either their actual shot or their vision system due to how low they were mounted.
Good strategy starts with your goals for the season. You can and should discuss this even before kickoff. Ours were as follows:
Be an alliance captain or first pick at all events
Win an event
Get to worlds
At kickoff, we decided the best strategy to achieve those goals would be to stay mostly on the opposite side of the field. We figured that at the district/DCMP levels there would usually be stray balls in the opponent’s sector from missed shots or unsuccessful human-player loading that we could pick up, so full HP cycles would be rare for us. In order to carry out this strategy, we decided we would need to:
Pick up from the ground
Score quickly and reliably in the 2 point goal
Avoid defense
Hang quickly every match
Our decision to go tall was based on these requirements. Pros for going high were:
Flatter shot meant wider range of shooting distances
Ensures our vision system can’t be blocked (there are unblockable shot trajectories for short robots but not unblockable camera angles)
Gives us the option to shoot from behind the control panel (which we didn’t end up doing but this made it at least an option)
Taller height limit makes climbing much easier
Easier packaging
Harder to defend
Don’t need a deployable mechanism to do the control panel (not that this really mattered in the end)
Better visibility on the other side of the field
Cons were:
Slower cycle times if we have to make full HP cycles
Harder auto routine to get 9 in auto (can’t get the two balls on the other side of the trench)
More likely to tip when crossing the rendezvous zone
When we weighed those pros and cons, it was a pretty easy decision for us to go tall. But if you had different goals or a different strategy planned, you could easily come up with different pros and cons that would lead you to go short.
I think you may want to reevaluate where you spend resources during build season next time. Increasing complexity in a drive system is, in my opinion, only (I say only with a grain of salt) worth the effort when testing something like a swerve in the offseason. For a fast paced game such as 2020, a butterfly drive gives you marginal returns on your investment. If you think about it, how often do you think you’d actually use the abilities of a butterfly drive during a real match? In what situations? How often would you actuate between the sets of wheels? Do those actuations warrant a full pneumatics system?
The slight maneuverability gaines is overshadowed heavily, again in my opinion, by increased complexity, increased weight, decreased real estate on your frame, which would better serve you in other subsystems that can have a much more calculable return on investment.
Sorry for going on a tangent but the offseason is a good time to reevaluate your decision making methodologies.
Design bandwidth, especially as the team hadn’t done a belt conveyor system in most of a decade
Fabrication ability, which was down a bit after one of our mentors moved for a job
Ability to package further control electronics in an incredibly tight electrical panel space (and that’s after already ditching Talon SRs and SPARKs for the smaller integrated-wire controllers)
Focus on reliable hanging using the Everybot climber
For us this year, yes. If we have the right kids this fall, I’d be open to CAD explorations of something that would clear the control panel. If we don’t, I won’t lose a wink of sleep over it.
I think we started to see that toward the end of the St. Joe event. Quick robots with high shooting percentage began to outscore. Of course in this game, you MUST also have a reliable climb. At the highest level, I think the triple balance will also need to be executed quickly.
The strongest points of our previous robots were their agility and speed, so we went for a low, light (<90lbs) robot, with the option to go under the control pannel, and as we went with a doble roller shooter, because we had done one in the off season, it would be harder to integrate it with a taller robot
