4 Cube auto Wed Practice match on Turing

Hope we can pull off a few more during qualifying matches on Thurs!

Oh wow… That is beautiful. Vision tracking the cubes? Or simply driving to their locations?

Looks great. Excited to see who hits 4 first in an official match. Maybe it’s 842 :smiley:

No such word can be uttered when talking about a 4 cube auto.

Our programmers will be jealous!

seriously… don’t show them.

Yes, I suppose “simply” is the wrong word, but I meant, is it turning X degree and driving X distance to get where the box should be. That is how we got our second cube, and I wanted to know if that same system was implemented on their robot as well, or if it was vision tracking.

No vision tracking is used. The robot is driving to their known locations.

Holy crap. That is absolutely incredible!

But… Is that really on a Divisional field?

I look at the stands and it seems a bit underwhelming for a Champs division. Not just the amount of people in the stands - but the size of the stands. Is this what FRC Champs is now?

Today was my first time in Houston, and while I will say I quite liked the load-in area and also the area outside the pits, the field area is quite underwhelming compared to St. Louis aesthetically-speaking.

That being said, the surrounding city and shorter walk to and from the pits are both nice. I have yet to form my complete opinion but so far I’m pretty neutral. It’ll be hard for Houston to beat the amazing memories I’ve had at STL.

absolutely mind boggling… congrats to your programming team! :eek:

What happens if the scale is up for the later cubes? Does not look like they can make it if the opponent possesses the scale.

If they can get 4 on that quickly, I’m not sure how an opponent would ever have time to tip the scale against them.

It is not turning X degrees and driving X distance to get where the box is. As you can see from the smooth motion, their robot is moving along a generated motion profile, which is faster and more precise than simply combining a turn + drive.

I really doubt that’s a problem.

If I remember correctly from last year, there are much larger stands behind the person taking the video. The stands on the other side are about 4-5 times larger than the stands that are seen in this video.

I was wondering how teams got their auto that smooth. We always put in degree and distance values, it has worked fairly well in recent years but it makes sense that motion profiling is faster.

Yep! It’s a commitment, but once you figure it out, it’s amazing useful.

I’ll admit that my perspective was similar to yours when I first got here, however since then it has changed for 2 reasons.

  1. I really enjoy having the fields and pits in the same area. The whole event just seems cohesive, and the shorter walks to and from pits/fields are a huge plus.

  2. As a spectator I vastly prefer Houston to STL. The stands are large and roomy (not that comfy, but what bleachers are?). But more importantly, they are right next to the field making it easier to see what’s going on.

I’ll note that this all applies only to the George R Brown Convention Center and division play, not Minute Maid Park (MMP) where Einstein will be. More than a few people on my team had complaints about MMP from last year including long lines to get through security, poor audio, and terrible viewing angles/distances.

You have to consider failure mode. You don’t just want your auto to hit all 4 cubes or miss all 4 cubes. Let’s say 842 places their pre-loaded cube and then misses their 2nd cube due to some sort of misalignment (it happens to everyone). It’s up against an opposing bot with a 3 cube auto who by that point already has two cubes up so 842 starts loosing the scale. Do you want the next 2 cubes to hit or miss? A slight tilt may seem like a small amount but it can mean a big difference when placing 3, 4 cubes.

Additionally like I was saying in another post, there are a number of robots that intentionally shoot their cube downwards onto the scale with significant force. This can offset the scale just enough to mess up an auto.

We’ve actually seen this in practice during auton scale battles in Ontario. There are robots that have an incredibly consistent 3 cube when their auton is uncontested but once contested their consistency goes down significantly. There are others that almost never hit all 3 when uncontested but become much more reliable when faced against a opposing 3 cube.

842 hit it on Turing Q54. Congrats on the world’s first successful 4 cube auto!!