It would have been a more interesting engineering challenge if that were true. Team 1827 will contend that minibot speed is actually all about eliminating every gram that isn’t mandated by the rules.
We’ve got a solid 2.5-second minibot without doing that, but every attempt we have made to do a direct-mount has resulted in a bot that does not have the torque to climb.
I know it’s possible because I’ve seen it done, we just can’t seem to figure out how to do it! (Though we’ll be spending a long Saturday tomorrow trying more ideas… We’d be working today if not for the Snowpocalypse.)
This is true, to a point. Once every gram is eliminated, how do you make it faster? If you have a few weeks until your competition, I’d suggest exploring how (subject to your capabilities and resources).
I may investigate a direct-drive option this weekend; though we also need to do a better “V” for the deployment mechanism.
As for minibot that takes less than 1 second to go > 7.5 feet? Well, I’ve had a bit of foot-in-mouth about minibots this season so I won’t really comment on it.
Alan is absolutely correct. You will understand it when you get there. This isn’t a meant to be insulting, just noting that it takes a lot of experimentation or very thorough design to not “luck” into a fast minibot.
Yes, weight is important, but so is frictions, traction (yes different), load distribution, and most important a matched gearing to the above parameters. When your gearing matches those, you get a very fast (maybe not 1 second, but fast none the less). We had 2 mini-bots of similar design, similar gearing, but one was significantly faster. Then we found and eliminated some scrub, and the slower one was significantly faster than the original fast one. After that, we got them both within 10% of each other with regards to run to run variability.
As JVN points out, iteration is often the key to incremental improvement. We tried approximately 5 “styles” of minibots before settling on an architecture that we liked. We have since gone through 7 different minibot chassis for this particular “style” of minibot. For those of you counting at home, that is around 12 minibots, and I hope to continue as i would like to revisit a completely different “style” that I am seeing get good results. It didn’t have good results in Week 5, so we put it on the shelf to be revisited at a later date.
Team 190 has quite the out of box approach to accomplishing the minibot task.
I recommend that come the WPI regional, watch match videos just to see it.
I am sworn to secrecy, but I will tell you this much, it’s pretty brilliant.
oops
The trick isn’t the motor, but what you use for traction.
There are teams out there that have to be working meticulously on this right now. They aren’t going to give up because of it. Trust me, there will be a sub 1 second minibot.
No way. Not including deployment.
I didn’t add deployment factor to the post did I. I’m strictly just talking about the minibot. Quickest minibot plus deployment will be around 1.75 seconds. Maybe not, if so, prove me wrong.
They will.
That would be tricky, since clipping to the pole requires turning a horizontal force to a vertical force, and you can’t leave too early after clipping because the minibot would bounce and slide around.
Our team bounced around an idea about a modified clay-pigeon launcher type thing that slams the robot onto the pole (lightly), but would deploy extremely fast.
We haven’t had time, money, or enough of a workforce to build it or the minibot, so if anyone wants to use the idea, feel free.
Actually it doesn’t…
Unfortunately I can’t say more.
I dont know if team 118 has said anything so i wont spoil it. But if you want to check it out go onto youtube and search team 118, it should be the 3rd video down. About 1 min and 49 secs long. Check it out and maybe you’ll be mind blown as much as i was. And i showed my whole team. All mind blown. Just saying, watch it, all of it, it’s got a secret but you wont know unless you watch the whole thing that i didnt do for 15 times until someone told me the secret and its great.
Inspired by that video, we worked for about nine hours today and managed to keep our current minibot consistent with our very consistent deployment apparatus, but cut the time from ~3.5 seconds to ~1.9. It was time well spent!
We just finished the Lake Superior Regional. We (2530 “Inconceivable”) were the first alliance pick of the #2 seed, 2169, King Tec. In the second elimination match we played, I believe we got as close as anyone to the best time (both time of Deploy+climb as well as clock time at end of climb).
We had a superfast deployment using preloaded surgical tubing (maybe <0.3 sec?), released by a cylinder under robot control at exactly the 10 second mark (its release had to be authorized by pressing a driver station button prior to 10 seconds). We also borrowed 2169’s minibot during elims (they brought at least five of them to the event). Theirs was a faster climber than ours, but not by all that much, ours never lost a race in all cases where it was successfully deployed (which was most of the time, not sure of the stats for attempts/success/fails)
I was the drive coach, and didn’t monitor the clocks for this, but I wouldn’t be surprised if hit the top before the clock read 8 seconds in that second elim match.
I never would have expected a borrowed minibot adapted to another hostbot would be so effective. (Sat lunch we very quickly made some relatively minor changes to our deployer to adapt to their minibot mounting design, though we retained our ‘turn on the minibot’ mechanism.
John Vriezen
FIRST Team 2530 Mentor and Drive Coach
Congrats on having a minibot that can get to the top of the tower within 8 seconds left on the clock!
I suspect that most regionals will have at least one or two teams with a minibot deployment system and climb speed capable of the feat.
In the Granite State Regional finals (Week 1), there were 4 successful minibot deployments – the last minibot got to the top with 6 seconds on the clock!
A super-slowmo video was put together by team 20 to confirm that the referees got their call correct for minibot ordering. (There is a separate thread discussing this video at http://www.chiefdelphi.com/forums/showthread.php?t=93314.)
One of the neat things about the video is when you view it at maximum resolution in full-screen mode, you can see the field clock at the far right of the blue alliance’s driver station. 1519’s robot hits the tower with 8 seconds left on the clock, with 40’s right behind at 7 seconds. The minibots of 175 and 20 both reach the top just as the clock flips to 6 seconds remaining, with 175 being only the slightest fraction of a second ahead of 20.
From what I saw of the WPI regional on the webcast, I think 40s deployment is faster now than it was at Granite State Regional, and they likely now reach the top with 8 seconds left on the clock as well.
At later regionals, more and more teams will have faster and faster minibots…
The real question is are we going to see a less than .75 second minibot.
I’d be a bit wary of automated deployment mechanisms. My team used one that fired at the 10 second mark, and we lost an elimination match when the ref said we broke the plane before the 10 second mark.
The cRio timer isn’t perfectly in sync with the field timer.
In the second round of the semifinals (which we would have lost anyway), we were flagged for deploying early, too…
Video vindicates us; the refs made a bad call.
It happens. No big deal…
…that said, everyone has already been cautioned against using a timer (or a light sensor) because the official clock is the only thing that matters.