4607 Reveal Video 2019

So we aren’t like @Michael_Corsetto and 1678 getting things done in the first try… and all that iterating has taken a lot of time! So instead of a cut and polished reveal video, this is what we *have.

4607 2019 Robot Reveal

*This is just the practice bot which is basically an ugly cousin of our comp bot. We’re running everything with open loop controls (manually) in this video and with no camera. Lots of iterating and improvements to be made!

15 Likes

Looks really good guys!

1 Like

Impressive cycling, especially for a week zero. With some tuning and more practice I could easily see you going far.

1 Like

Nice!

Love it

1 Like

@dydx, my thoughts exactly. For week 0, impressive!
More practice and tuning will only make you better.

With some extra practice you can realistically solo a rocket under no defense. With partner help on the low level you could get the extra RP most matches. Nice work!

1 Like

Yeah this was only the 2nd day we had a functional robot… got everything together and semi-functional at 2AM on Saturday morning. Probably only have 3-4 hours of drive time before this video was taken. We were also messing around with the manipulator design and stumbled across something that worked extremely well by dumb luck (or it could be that we tried everything else already). With that said, I think there’s a crazy amount of room to improve, especially when the Limelight get plugged in and our control loops get tuned.

With regard to the Rocket RP… I think that it’ll be tough. If a team wants to stop you from getting it they will. Defense is brutally effective from what I’ve seen. The best defenders are the teams on your own alliance though… sharing the human player stations and navigating through traffic slows things down significantly.

1 Like

Here’s another video with a lot more mistakes if anybody is interested.

3 Likes

Great cycle speed, we too are working on the solo rocket RP and wanted your opinion. With 3 pts for cargo, have you considered going back and forth between cargo and hatches filling one level at a time to maximize your scoring or is it faster to focus on one game element at a time?

We definitely tried both and we found that we liked getting into a rhythm with one game piece and then going to the other. The same was true for filling the back of the Rocket and then the front. It’s just nice to do the same motion over and over rather than switch between them every time. On the other hand when Cargo spills out in front of the Human Player Station, we want to just intake it and score it rather than try to knock it out of the way.

So the strategy we came to by the end of the day was essentially fill the bottom two with Hatch Panels (to enable us to grab and score Cargo if it’s in our way without needing to worry about it falling out) and then follow that up by filling the remaining Hatch Panels on the back, then the front, then Cargo from bottom up.

The disadvantage to our strategy was if you make mistakes (or are otherwise too slow to fill the Rocket), you’re leaving points on the table by spending more time on Hatch Panels rather than Cargo. Our theory was essentially that if you’re going to be nearly soloing the Rocket, you’re probably going to be winning the match either way… so missing out on a few points isn’t a huge deal. Our strategy would obviously change to maximize total point output in close matches, or matches where we’re the underdog.

The other thing we found was that intaking Cargo from the Human Player station was extremely reliable and difficult to screw up while intaking from the Depot was slightly faster but more prone to mistakes. What we eventually settled on was going to the Depot for our low Cargo cycles, and going to the human player station for mid and high cycles. This helped us to minimize our elevator travel in order to save some battery and time. I think that will be less of a consideration on the comp bot with a more efficient and faster elevator.

Hope that helps! I’d love to answer any other questions that anybody has!

3 Likes

6 hatch panels and 4 fuel, nice!

Both matches are fantastic!

Something doesn’t seem right here :thinking:

4 Likes

They’re round and an obnoxious color. I don’t see any problem.

2 Likes

And they get in the way of driving. If it looks like a duck and talks like a duck…

1 Like

With the right bumper material the cargo is no problem. But the right material sucks in every other aspect.

Not gonna lie, when I posted the reveal video here and submitted it to FIRST’s Build Season Celebration show I completely forgot about all my stupid commentary… “care-go”, “RETRACT!”, “LEFT, DOWN, ACTUATE!” Whoops.

I’m used to looking stupid in front of large audiences though so it’s all good.

4 Likes

It depends on multiple factors. First, you should consider the types of intakes and speed to intake and place hatches and cargo. You should also factor in where you are on the field. You should also take into account your bot’s drivetrain. If you have a faster intake for a certain game piece than other bots on your alliance, probably aim for that game piece, but also factor in where the rest of your alliance members are and routes they may take to put their game pieces in as well as how far away certain game pieces are from bots in your alliance. You might want to also think about the capabilities will probably make a difference, too, like if you have a holonomic drivetrain with a hatch closer than a cargo (go for the hatch) or a tank with a cargo 2 feet to your right but a hatch 5 feet in front (go for the hatch again). Don’t forget to think about the abilities of your intake to pick up game pieces in various positions as well as the positions of other bot on your alliance (if you’re good at cargo and another bot is picking up a cargo and gunning for the level 2 rocket, both your bot and the last bot should pick up a hatch and place it on the rocket simply because of efficiency). There are other factors to consider, but these are all I could think of at the moment.

So finally, here is a picture of our 2019 competition robot: “Actuate”

View post on imgur.com