I have heard quite a bit both ways can people add from experience with your own stations or actual field items which you believe to be true. Perhaps if someone could help us understand why some seem to work and some don’t. Just construction?
We aren’t counting on them landing flat, in our testing it’s been <10% landing flat. As to why, I think it’s just the slow speed at which the totes exit the feeder station. If they exited faster, such as when pushed (illegal) they land flat nearly every time.
On edge seems to be most common for us. With our current design, it is going to be a big time suck to have the bot right it before picking it up. I hope teams are considering this in their strategy.
We experimented with the field in Manchester and got a consistent result of tote bins landing on end as they came out of the feeder station. Our theory is that the home version of the feeder stations teams are making do not accurately simulate the constraints the feeder station ramp, guards, and lexan edging that constrict the tote. We are going to tweak our home version until we get the bins coming out like the field but for now they are landing flat on the floor which we know is not accurate.
Tote bins with no interaction from a robot take a nose dive as they come out and stand up on end.
Our testing showed that totes tended to most frequently land on their side after coming out of the tote chute. Occasionally we would get one to land right-side-up, but the totes generally came out slow enough that the friction with the carpet could “grab” the corner and tip it up on end. If the totes came out faster they would land flat more frequently.
Do not expect totes to land on their bottoms. If you want to use totes from the humans, you need to have a way to knock them over or pick them up and use them that way.
we need a robot to assist with the exit of the tote from the chute
This kickoff video had a official feeder station fast forward to 57seconds https://www.youtube.com/watch?v=Y1-3u9NEeO4
One way that we found that we could get it to land flat was if we released it when the tote was high up on the ramp(don’t think this counts as pushing) and quickly open the gate before the tote reached the bottom. It takes a lot of timing and practice, but it allowed us to make the tote go fast enough to land flat.
Totes generally (but not always) land flat from our chute.
For what its worth, our chute has an HDPE base with aluminum angle for the guides. The enclosure is plywood. The height and angle are reportedly correct, but I have not personally measured/verified them. At the moment, it does not have a regulation gate/door - we are using a hand-held piece of wood. We will have a real gate soon. We plan to raise/lower our chute to see how that affects the way the tote lands.
It seems to me that the chute design puts performance right on the cusp between totes landing on end vs. flat. Small variations in chute height, base material, carpet, etc., may have a big effect on results. Has anyone done any reasearch/testing to determine what factors affect results any why? There seems to be a lot of variation between teams’ results. Can we expect similar variation on the real fields?
If you have responded to this poll/thread, please describe your chute construction, etc., along with your results.
(Bottom line, don’t count on totes landing consistently)
Team 3081 did extensive testing with the use of crates to get crates to drop flat. You can take an upside down crate, and push it wide ways up to the chute. The crate would almost always land flat.
The following Q&A may be of interest:
I agree with Wayne that the chute design is on the cusp between landing on end and landing flat, with a slight bias towards landing on end. But, I expect this to vary somewhat from chute to chute.
Our chute is as close as possible to the official design (correct height and angle, HDPE floor, aluminum angle, lexan door, being operated on carpet) and our totes nearly always land on end.
If you expect all totes to land flat, you’re gonna have a bad time.
From some testing, our totes land mostly on-end.
BUT! If you plan on Catching totes, anything taller than say 4 inches will catch a tote flat.
I was just reading Q&A on my lunch break and noticed this (Q&A 127). Good enough evidence to convince me that the ambiguous landing orientation is intentional. In fact they probably put a lot of effort into making it that way.
We made a “team version” tote chute. Totes slid very slow and landed on forward end.
We added smooth plastic to the chute, like in the actual field, and they slid faster and bounced onto their bottom. Though they were never straight.
To those teams that have made their chutes more closely resemble the official FIRST version: What is your confidence level that the human player can make a stack of 2 totes that is either nested or close enough to call it an effective 2 stacking option?
We played around with this on the official field. You won’t be able to make a two stack without robot interaction. The first tote always takes a nose dive and lands straight up. Once a robot gets it upright in front of the feeder the human player can send another down the chute and it makes a pretty decent two stack but it isn’t perfect.
Try this…Load the Tote into the Chute, release the Tote, open the door handle, and as soon as the leading Tote edge exits the field side of the door edge of the chute, release the door back down onto the Tote, it may hold it up on the rear long enough during the sliding exit that the free end does not dive as much (and land on the Tote end that often), and therefore, lands so that the bottom of the Tote is flat on the carpet a bit more often.
It appears from the videos I have seen of actual field Tote chutes in use (testing), that holding the door up too long actually inhibits it from working as originally designed. Go back to the video of the actual field elements being used during the game release and watch it very closely. (Then practice, practice, practice).
This hasn’t been our experience actually. We have seen about 80% landing on their bottom. Our difference though is that the chute bottom is made out of polycarbonate, not HDPE. Everything else is very close to the field specification. We are trying to figure out what is causing our results to be so different.