Team 1658 3 ball autonomous routine in 8 seconds(practice bot)

Will add function to move forward as well and potentially hot goal targeting. As it is now we will always get at least 1 ball and 2 if we are lucky but may try to squeeze out another 5-10 points for maximum pointage if we can hit hot goals every time

Video or it didn’t happen.

oh i’m sorry

Did not expect that. When everything popped out I was rather surprised. Nice work!

That’s actually thoroughly fantastic, and one of the most unique mechanisms I’ve seen this year. Really cool.

Good job!!! That is so cool guys. I love your guys technique for the triple autonomous.:cool:

I call it the spread eagle though i’m not sure how appropriate that name is… lol

ye we tend to come up with strange ideas. we probably use the strangest drive as well. We are just a strange team i suppose =p it seems overly complicated with about a hundred parts but hey it’s super effective and designed to be repairable within just a few minutes so we should be fine… i hope

uhhhh… :yikes:

now i just gotta make it move for 5 more points and maybe add hot goal detection for another 5-10 points. then it will be as good of an autonomous as theoretically possible for a single robot. the 2 remaining seconds should be enough time for that right… right? xD

I believe! This is enough to win most matches on its own already.

Although, I’d focus on making a perfect two ball auton. I think it’s safe to assume that you will have a partner in elims that can do a single ball auto.[/quote]



Well your partner in elims might be a defensive or goalie bot(rarely). And even if it is elims having a triple auto work like it did in that video, would get them a lot of points.[/quote]

you are thinking too linearly, there is plenty of time to move forward and do vision tracking. these functions don’t have to be serialized they can all happen at the same time. the only thing required is that when a ball is actually being shot forward that the robot not move, otherwise the other tasks can happen at the same time as the rest of the routine meaning it won’t require any additional time and we should still have about a 2 second margin, even if it takes a little longer i think 20% margin(2/10 seconds) is enough time, if just barely[/quote]



that was the idea from the start, is maximize auto points, do 2-4 balls during teleop with 2 or 3 assist and we should average around 200 points a game hopefully. we did 3 ball autonomous because we believe we can make it work 100% every time so even if another robot can do it unless they are confident it will go in 100% we want to be able to do it ourselves and be 100% sure. can’t let those points go to waste.[/quote]


it should work like that every time. everything on this robot is extremely forgiving, precision not required. and mid development when everything wasn’t tweaked yet worst case scenario was it was still making 2/3. the ball we possess will always go in we know that. and if one side screws up for whatever reason the other side should still work. But we are trying very hard to make it work every time reliably.[/quote]


Well post a video of you guys doing triple auto, moving forward, and having vision tracking all in ten seconds and then I will believe. I think you guys can move forward but having vision tracking all depends on if the hot goal lights up on the same side you are on during auto.:rolleyes: You know maybe im wrong and over looking something but i will stand by my opinion and we will have to find out.:cool:[/quote]

Very impressive autonomous!! I’m sure 1658 will be extremly competitive at the St. Louis regional this year again. The only thing I worry about is defense breaking your robot. After watching many many matches I beleve that defense is going to be brutal this year and when a robot collides with your robot at 10-15 fps bad things can happen!!

That autonomous though is just epic…EPIC


the code for this was just done today so with a bit more time i’ll get the rest in. figured it was good enough for a post though. we’re going to have an ice storm next two days so those are shut down but expect a video of at least added movement sometime next week. Also we will be in the smack middle of the field anyway. it will require either a slight left or slight right turn to make it but that’s desired. This way with a simple 20 degree rotation we’ll be facing the other goal, it’s fast and simple and doesn’t have to be accurate either.

i’m still trying to decide what would be an easier way to program the detection. We can also use a kinect and detect which hand of the driver is up. you are still allowed to do this no? we know the goal switches at 5 seconds so the only detection that needs to be done is a single frame at the very start which really is no big deal anyway. We just always had issues with the field lighting screwing up everything but i think I can figure something out this year if need be. But is it worth it for a measly extra 5-10 points though? you’ll be hearing more from me later =p[/quote]

we expected this and specifically designed around this. there is only one time we are ever vulnerable to breakage, and that’s the moments between putting the roller down to pick up a ball and the time it takes to get the ball in the robot. However due to our agility even the risk during this period can be minimized by the driver since we can rotate on a dime and move at the same time. otherwise we are completely safe from breakage. But you may notice how these things are made. There is nothing that extends outside the profile that can’t be replaced within 30s-1minute. In fact we are counting on having to fix everything after every match and we are prepared for just this scenario. since we have two picker uppers anyway we can be risky until one breaks, then just use the other. but even ontop of all of this, almost everything that extends out can break, and we can still play. we may break easily, but just because we break doesn’t mean we are out of the game. far from it actually.