ball cannon???

[stinkypooman]

Im 100% sure were not the only team that had a debate over this design. Quite frankly at first i thought this idea was pretty dumb... jeez its hard enough for a person to get that ball in, how the heck is a robot gonna do it, espically on that crazy surface with the crazy wheels. But Ive been thinking a little more now, if you position your "ball cannon" higher than the low edge of the trailer and make some kinda of rotating turret stand for it then mabye it could be effecent from distances of 5-15 feet if you chased robots from behind. It would be a programming nigtmare though. I guess we would have to mount the camera so that it can track the colors on the trailer and look at the angle and from there determine the distance. but heck it was our programmer that made up the idea... i mean that doesnt solve the problem of the skidding surface or the lag made form the air time fo the ball...

any takes on this? pros cons?

[Darkcrosbone]

erm.... in my opinion i think launching is out of the question.... better just to throw from the stations and to sort of dump the balls into the other alliances trailer

Edit: tho i AM a rookie, anyone more experienced want to comment?

[ginosoprano09]

I was actually going to start a thread about this so that more experienced teams could make some suggestions about it.

That's the design I want to use,but I'm not really sure how to actually build one.
(Anyone want to take a shot at it and describe how to build one? )


erm.... in my opinion i think launching is out of the question.... better just to throw from the stations and to sort of dump the balls into the other alliances trailer

Edit: tho i AM a rookie, anyone more experienced want to comment?


Relying on dumping and throwing will be less efficient though. In order to dump the balls on another robot, both robots have to come to an almost-complete stop, and considering how slippery the surface is, both robots are going to end up skidding in a small area and will make them easy targets for the throwers.

You cannot rely on your throwers because teams are going to have solid robots that are going to find an efficient way to overcome the slippery surface and will be a fast moving target. And as we all know, nobody counts with a perfect aim.

So does anyone know how to build a ball cannon?
( I might start a new thread on this)

[Taylor]

For those of you who are relatively new to FIRST, in 2006 there was a game "Aim High" where we were asked to launch 7" poof balls into a vertical target eight feet off the floor. Many teams used a modified softball pitching machine design for their robots, with moderate to high success. Although the geometries of the poof balls and the orbit balls are quite different, this could be a possible strategy to revisit.

[stinkypooman]

Originally Posted by Taylor
For those of you who are relatively new to FIRST, in 2006 there was a game "Aim High" where we were asked to launch 7" poof balls into a vertical target eight feet off the floor. Many teams used a modified softball pitching machine design for their robots, with moderate to high success. Although the geometries of the poof balls and the orbit balls are quite different, this could be a possible strategy to revisit.


this was actually what we were thinking about using like the tennis/baseball/softball launcher type design. im still a littel skeptical beuacse we can argue that its going to be hard to use a dumping design accurate, but that doesnt change that fact that a launching design would be any more accurate...all of this is super dependent on how much sliding we get on the new surface because really no one knows how different its gonna be from carpet. also the balls are not soccer balls anymore, they are plastic wrapped in cloth, this years balls will definitely have more give than soccer balls used back in 2006. This may make the "baseball machine" design useless or highly ineffiecent? i dont know, maybe it will work? :confused: I was thining maybe a pneumatic cannon thats nice and simple, but the problem with that is the velocity will not be variable like the baseball pitcher design would.

[Raumiester2010]

I was actually going to start a thread about this so that more experienced teams could make some suggestions about it.

That's the design I want to use,but I'm not really sure how to actually build one.
(Anyone want to take a shot at it and describe how to build one? )





Relying on dumping and throwing will be less efficient though. In order to dump the balls on another robot, both robots have to come to an almost-complete stop, and considering how slippery the surface is, both robots are going to end up skidding in a small area and will make them easy targets for the throwers.

You cannot rely on your throwers because teams are going to have solid robots that are going to find an efficient way to overcome the slippery surface and will be a fast moving target. And as we all know, nobody counts with a perfect aim.

So does anyone know how to build a ball cannon?
( I might start a new thread on this)


Our team debated about a dumping mechanism too, we elected not to use it because in order for it to work, you have to get up close, be fast, accurate, and (this is the hard part) find [I]some[I] way to get it in the opponents trailer without leaving your 28" X 38" X 60" box. (remember that rule, we hate it too).

That is why we are planning on some sort of shooter. haven't made too many decisions about that though.

[Ryan Albright]

I am not sure I am feeling the Launcher. I think the most efficient designs are going to be a robot that can suck up the balls and place them in a hopper so they can be dumped or a robot that can "plow" the balls into the corral for the human player. Over the years I have been involved these kind of robots always succeed in the end.

[XXShadowXX]

I think the most efficient robot will be one that can not only scope up balls, and dump them into nearby trailers but...
track oposing trailers autonoumly, and launch balls across the feild into said trailers

[sdcantrell56]

I know we are fully planning on a fully turreting ball launcher. We will also attempt to track the opposing trailers at all times and have the computer handle all the shooting. If no one has tried it yet, it is quite easy to make the balls in the trailer. We are also planning a very quick ball harvester and large hopper so we can easily corral balls as well. I don't think launching balls from any distance will be easy, but some teams will undoubtedly be able to accomplish this task, and the best of these will be the champions. This is of course just my opinion.

[Petey]

My $.02? You will see almost no successful launchers this year. A few teams will pull it off, and we will marvel at them and slap our forheads and say OF COURSE, but most of the most successful teams will find a way to get balls reliably and quickly to the human players (including all four empty cells) and then have a human player who is absolutely golden scoring goals.

Someone namedropped Aim High; I'm thinking more like 2004 here.

[Daniel_LaFleur]

My $.02? You will see almost no successful launchers this year. A few teams will pull it off, and we will marvel at them and slap our forheads and say OF COURSE, but most of the most successful teams will find a way to get balls reliably and quickly to the human players (including all four empty cells) and then have a human player who is absolutely golden scoring goals.

Someone namedropped Aim High; I'm thinking more like 2004 here.

I seem to remember people saying this about hurdling last year, too.

[Enigma's puzzle]

I seee the teams that have been around a lot longer will have an advantage and will be more likely to build a succcessful shooter, EXCEPT there will be a few teams that find and make simple shooters. the rest of us that didn't attempt it because of complexity will hit ourselves.

And of course thier will be a fair amount of unsuccessful failures.

[MrForbes]

I know we are fully planning on a fully turreting ball launcher. We will also attempt to track the opposing trailers at all times and have the computer handle all the shooting. ..... We are also planning a very quick ball harvester and large hopper so we can easily corral balls as well.

It's good to see we're not the only ambitious 4th year team :)

[BigWhiteYeti]

The balls are a little irregular, so using a flywheel cannon would be a bit inconsistent from one shot to the next. Plus, to score on a shot ball, you would have to miss the center pole and hit just on the side, including the correct height. It would be extremely difficult to do this in my mind.

[sayso_411]

For a shooter to work you need to stay under 60 in. at all times. Saying that you have to top feed the shooter and have the shooter at the bottom. Teams who are thinking of big hopper and a shooter will a problem if they are attempting a rotating turret because of the height restrictions. say 20" for a hopper that leaves 40" (the optimal height u want ur shooter to be.

I see a problem...

good luck

[Dominicano0519]

think one thing wiffle ball because this ball is just like a wiffle ball its really hard to control the movement of the ball and its so light it will be hard to throw very far

[sdcantrell56]

I definitely have an idea for at least a 40" tall hopper, basically the dimensions of the drivetrain with a turreted shooter at 40"+ I am not planning on posting up the design or anything but so far its looking like it can be done fairly easily with room for at least 20 balls. Also throwing the balls is quite simple, at least for my team. :cool:

[=Martin=Taylor=]

I definitely have an idea for at least a 40" tall hopper, basically the dimensions of the drivetrain with a turreted shooter at 40"+ I am not planning on posting up the design or anything but so far its looking like it can be done fairly easily with room for at least 20 balls. Also throwing the balls is quite simple, at least for my team. :cool:

I assume a suction device on the bottom for added traction is a given?

How much force did you guys get out of that sucker in '08 anyhow? :)

[sdcantrell56]

Of course. Last year we had around 400lbs of force. We might just suck again this year :cool:

[JesseK]

It's good to see we're not the only ambitious 4th year team :)

Heck yea, we're doing a shooter too. I prototyped a successful launcher yesterday that went 8 feet, and the students are almost done prototyping a shooter today. Combine it with a turret and camera and you're golden -- they key is the type of shooter. Get creative, you don't have to use flywheel style...

We're not using a hopper though. Our overall game strategy dictates that we won't have the time to collect large amounts of moonrocks, so we'll fire what we get almost as soon as we get it. We'll have a max of 6 moonrocks in our bot at once, and probably then only at the start. This is for simplicity's sake though...

[Luke Pike]

I don't pretend to know much about this, but wouldn't all the holes in the ball make suction difficult?

[Alex Dinsmoor]

I don't pretend to know much about this, but wouldn't all the holes in the ball make suction difficult?

More like impossible :)

[Jon Stratis]

I don't pretend to know much about this, but wouldn't all the holes in the ball make suction difficult?

I know the location of a few wind tunnels that could probably do a fair job at providing enough suction to suck up a ball... but you aren't going to be able to get those under the weight limit, or even run them on a simple 12V battery.

So for all intents and purposes, suction on the entire ball can easily be considered impossible within the scope of the competition. However, a robot could *theoretically* suction a small part of a ball - one rib - and provide enough force to grip the ball. That is, if you have a mechanism that is sophisticated enough to be able to target and approach a single rib on a ball.

[EricH]

For a shooter to work you need to stay under 60 in. at all times. Saying that you have to top feed the shooter and have the shooter at the bottom. Teams who are thinking of big hopper and a shooter will a problem if they are attempting a rotating turret because of the height restrictions. say 20" for a hopper that leaves 40" (the optimal height u want ur shooter to be.

I see a problem...

good luckYou weren't around in 2006, were you?

[Raumiester2010]

Heck yea, we're doing a shooter too. I prototyped a successful launcher yesterday that went 8 feet, and the students are almost done prototyping a shooter today. Combine it with a turret and camera and you're golden -- they key is the type of shooter. Get creative, you don't have to use flywheel style...

We're not using a hopper though. Our overall game strategy dictates that we won't have the time to collect large amounts of moonrocks, so we'll fire what we get almost as soon as we get it. We'll have a max of 6 moonrocks in our bot at once, and probably then only at the start. This is for simplicity's sake though...

lucky, you guys get to start prototyping already? we are sitting around tossing around ideas... maybe today....

my $.02 (ok the second ones) Ill agree that the dang irregular shape will make it interesting to throw, we where planning along the lines of a few spinning wheels (kinda like in '06) but we keep thinking of the issue of the wheals getting stuck in the gaps...

[The_M01e]

In my opinion, a "pitching machine" is the way to go. Though the MoonRocks may look different than the Nerf Balls we used in Aim High, they still have enough shape to act like a ball. I think trying to pin down a robot and dump the MRs out and into the trailer will be a pain.

I think we're going to use the camera and tracking to hit our mark.


Hopefully that physics we learned about will pay off!

[CraigHickman]

In my opinion, a "pitching machine" is the way to go. Though the MoonRocks may look different than the Nerf Balls we used in Aim High, they still have enough shape to act like a ball. I think trying to pin down a robot and dump the MRs out and into the trailer will be a pain.

...Why not do both? Be able to pin em down and dump into their trailer, or just rip shots off at them from across the field. I predict the robot that does both well will win this.

[ginosoprano09]

Would it be legal to have a ball cannon that'll shoot backwards in order to block other balls??

It may difficult to coordinate the "collision",but is it even legal trying?

[waialua359]

I think shooters are possible and you WILL see a lot of teams doing it this year.
The most effective of the teams trying to do so will most definitely be using the camera to do this. FIRST has done everything to encourage this with the inclusion of a camera/lazy susan in the kit and flourescent colors on the trailers.
Go figure. ;)

[Team 1708 Dave]

We were also considering a "launcher" type mech. It would be hard to aim but using a dumper type you'll loose all your game pieces at once.

The use of a "harvestor" type ball collector.. is this even legal? Read the rules a few times and it really dosn't say either way except that you can't herd game pieces.

[EricH]

The use of a "harvestor" type ball collector.. is this even legal? Read the rules a few times and it really dosn't say either way except that you can't herd game pieces.Wait, what? Please, don't be so general. You can't herd Empty Cells if you already have an Empty Cell being herded or in possession (<G24>), but that's about it.

[KarlSTA]

i think they were referring to the legality about danger of wrecking the balls....not sure though

[Team 1708 Dave]

Ok.. you got me.. Sorry a lil tired, bear with me here. Not sure if it's legal to gather game pieces for the purpose of reloading your dumper or shooter during the game after you dump or shoot your 7 that you can have on board.

[sdcantrell56]

The suction we were referencing was in relation to suction on the ground to increase our normal force and thus traction. For anyone who is new, last year we had a suction device to hold a launch the balls that could grip with over 400lbs of force. That is why he was asking us about doing suction again. We will definitely not be attempting to suck the fuel cells this year

[jimbot]

Our team had this discussion the other day, and we concluded upon that a shooter will be very, very, very difficult. one piece of advice is when you (the human) shoot the ball at the trailer are you using a arcing basketball shot or a bee-line baseball throw. If you are doing the arc shot, try and score with a straight throw, because I don't know if you would be able to replicate a arc shot, let alone hit any object at multiple distances.

[scottyh2006]

well the lowest post are 32 inches so we are having a cannon bout 38" high and kinda just let the moon rocks and the cells just like come out not at a high speed just enough to make it go 1-3 feet.

[lingomaniac88]

My main worry about using a launcher/shooter is whether or not it will break the game pieces.

And Team 1708 Dave, it is perfectly legal to harvest game pieces off the floor in the middle of a match.

[ehcualp]

My main worry about using a launcher/shooter is whether or not it will break the game pieces.


if your launcher is capable of breaking the game pieces, it should probably should not be used. :D
My team is certainly going to attempt a launcher. It will be difficult to fire in the same spot twice, but after maybe a week or two of practice we could have it down to an exact science.

I seem to remember people saying this about hurdling last year, too.

I seem to recall hearing that too. And yet, nearly every robot at nationals had a hurdling robot. Go figure.

[GGCO]

Well after a thoroughly healthy discussion and brainstorming session tonight I have a few thoughts.

A basket idea is nice, but I believe that it is essential to have a system that can harvest balls (efficiently) from the floor of the arena and this is one of the things that is sacrificed by implementing a basket system that can dump left/right/forward. Also, with a basket I think that it is only worth it if it can hold large quantities of balls. So if it holds 5 and 3 go in and 2 miss, no big deal. But the thing is, that is the only time you can shoot so it (assuming that you have no harvesting system) so it better be perfect.

I think that if anything a herding bot should have some kind of offensive system (a basket would be the best), so when they get into a "sticky situation" they can maybe use it.


I think that a baseball type shooter is alright. There are some definite issues of how to get the ball into the elevator that feeds it. It really has to be a perfect scenario for a ball to find its way to the opening of the bot, not run away from it because it hit a part of the bumper, and get wedged into it and go up into the elevator. However, if you can get around this problem it's a decent system. This is because you can (potentially if your turret can rotate) drop/shoot balls into an enemy bot's trailer from all 3 sides of your robot! You can also control how many balls go into the trailer!

Lastly, I wanted to throw out another idea to the CD crow that I brought up tonight. I don't like it as much as a shooter, but maybe some team can get inspired by it.

I thought of having a robot whose front part of it's chassis is indented. This allows a place for balls to gather. A basket would scoop down and gather the balls. This would make the basket argument much more compelling, but it also raises a whole other set of problems. For instance you sacrifice precious space of your bot to accommodate the scooper, and I believe that a robot with a shorter width is very much essential for turning this year.

That's my $0.02!

p.s. I really had fun with BOLDING, underlining, and italicizing

[Team 1708 Dave]

Thank You Lingo! Did you find the section that approves this action or are you reading betwwen the lines like we are?

[ehcualp]

I think that a baseball type shooter is alright. There are some definite issues of how to get the ball into the elevator that feeds it. It really has to be a perfect scenario for a ball to find its way to the opening of the bot, not run away from it because it hit a part of the bumper, and get wedged into it and go up into the elevator. However, if you can get around this problem it's a decent system. This is because you can (potentially if your turret can rotate) drop/shoot balls into an enemy bot's trailer from all 3 sides of your robot! You can also control how many balls go into the trailer!

My team has a great idea for grabbing balls from the ground, and having up to 10 waiting to be fired with the baseball shooter. I'm 100% that there will be problems with it, but it will be awesome if it works just like we plan it to.

[sxysweed]

Our team is... Broke. And we have very little options as far as welding and machining... But we've been tossing around the cannon idea. Pick the balls up and hold them monetarily, but have the cannon able to rotate (via control from the drive) but shoot at a fixed angle, variable speed. Kinda difficult, but managable. Conveyor belts used to shoot it out as speed.

First thing I thought was the ride-up-and-dump thing. But the two main things with this are a.) Bumper Rule: 60% must be covered and b.) slipery floors make this difficult.

[GGCO]

I think that if your turret doesn't rotate it isn't worth doing.

Think about it...

If your turret doesn't rotate you have to have the robot directly in front of you. This would be a big challenge if it were regular flooring. But it's not regular! The flooring will make it near-impossible to have an effective fixed-turret system. However, since your team is short of money it is better than nothing, unless you can get creative....

[lingomaniac88]

Thank You Lingo! Did you find the section that approves this action or are you reading betwwen the lines like we are?
I didn't see anything in the manual that explicitly allows this, but I noticed that one of the robots in the game animation is able to capture game pieces on the floor. So I don't see why this can't be done.

I think conveyor belts are a really good way to manipulate the game pieces. My team used conveyor belts in 2006 for a floor shooter, and from what I heard, it worked very well. I think this could be applied to shooting or harvesting game pieces, as well.

[gorrilla]

maybe....if only you could somehow combine the harvester, with the shooter... into a single mechanism:rolleyes: (hint hint)

[rcmolloy]

It seems as if you need an athletic team to be with this competition. I played basketball for 5 years so that I give me if I am a payload specialist a little edge. I think that it's going to be very different from what has been previously played.

[elvis9110]

i think that a cannon that tracks with the camera and stays on target, but is also controlled, at least to some degree by the driver would be the best cannon option. i know this would be bad for programmming.

[Uberbots]

This game was practically designed to be a shooter game. It was hinted to man a times in the kickoff that robot to robot contact (or robot to trailer for the purposes of scoring) would be very difficult to do and would require much time and coordination (and i don't believe the other team would be so keen about you scoring in their trailer)

The camera systems given to us along with the lazy suzan (and even the game video) is almost a direct notice to us: use a shooter! but the question is how you would make it effective?

unlike 2006 where you would fire at a stationary target, this year you must fire at multiple moving targets which you cant assume will be in front of or behind you. it would be near impossible for a pilot to aim the turret at an opposing robot, accounting for all the possible motion and angle, and fire a successful shot. We are, however, armed with an extremely powerful vision system this year, which can do advanced color and blob analysis at speeds inconceivable with the CMUcams.

The basic logic i have hammered out is you would have a camera mounted on a pan/tilt on top of your turret. This will give your cam a constant, known height. we are also assuming you have a well designed shooter with a control loop that can have an accurate firing distance and angle, as well as approaching those parameters in under a second. The camera is on the constant hunt for beacons that your team wants to shoot at. when it locates a beacon, there is a control loop that tells the camera to lock its pan/tilt such that the beacon is in the center(ish) of the feild of view. Servo/image error will then allow for the angle of the turret to be corrected for.

The interesting part comes with shooting. Both you and your target are (perpetually?) moving. to account for this, your software will need to be able to do motion prediction (probably using second-degree extrapolation of some sort), and using empirically or mathematically determined data, will be able to lock onto where the target will be. With a system like this in place, even a fast moving and evasive target will not be able to outsmart your turret... well unless its out of range or they move in a manner that second-degree extrapolation cannot predict.

Software is incredibly important this year! you can close your loops with far greater precision (double precision floating point values!!!!!) than you could in previous years.

[midway78224]

our team was thinking about that today and we decide to try to build a harvester and turret for the balls. We are always going to try to make the auto detect thing work with the camera. Were going to start testing that theory later this week and see if it works.

[JM987]

Ok.. you got me.. Sorry a lil tired, bear with me here. Not sure if it's legal to gather game pieces for the purpose of reloading your dumper or shooter during the game after you dump or shoot your 7 that you can have on board.


No, you can only have one empty cell in possession of your robot, but the number of moon rocks is limited to however big the room in your ball holding mechanism is:)

[manderson5192]

Of course. Last year we had around 400lbs of force. We might just suck again this year :cool:

Ok. I have to ask. I, too, have looked into using a vacuum to increase traction but I have run across a seemingly-insurmountable roadblock to the plot and it involves <R06>.

I had envisioned a tube or suction cup of sorts that would need to be pointed at the floor with a rubber shroud going around the edges that would just barely clear the ground. Here are my concerns though: 1) How do I make sure the shroud doesn't touch the ground, thereby violating <R06>, and 2) Will this even work to a good extent? Will there be too great of a loss of vacuum?

Additionally, I envision that the fan in the vacuum generator would suck air from under the robot and blow it out above it. Am I right in making this assumption?

Thanks!
-Matt

[ScottX]

True! It would take so much time...and a master programmer...x]. My team was thinking of using a cannon also...i don't think its a good idea though...since with the robot sliding..=/..and a cannon that isn't that accurate..=/..geez how will this work together...><! HELP!..=D

Pros: Robot gets to shoot!

Cons: Will it work? Is it accurate?

[ehcualp]

It all depends upon the driver. IF that driver can successfully figure out how to drive your robot the way you want it on the slippery surface, then it doesn't really matter what launching/dumping device you use.


This competition most relies on:
1) a great driver with lots of practice and experience
2) someone on a basketball team
3) a good strategy involving your two alliance members.

[Jon Stratis]

True! It would take so much time...and a master programmer...x]. My team was thinking of using a cannon also...i don't think its a good idea though...since with the robot sliding..=/..and a cannon that isn't that accurate..=/..geez how will this work together...><! HELP!..=D

Pros: Robot gets to shoot!

Cons: Will it work? Is it accurate?

Actually, the robot sliding helps a shooter design, not hurt. Due to the smooth surface and the CoF of the wheels, teams are going to be finding it very difficult to turn on the surface - once they start in a direction, they are going to keep sliding in that same direction for a while (a few seconds, a least). That means that a shooter system can assume a straight line travel vector, and compensate for the air time of the ball. If we were playing on carpet with last years wheels and robots, a shooter system would have no hope of hitting a target, as the robots would be turning and moving around too quickly.

I would say the playing surface is the only thing that makes a shooter system viable this year.

[Animator]

i think that a cannon that tracks with the camera and stays on target, but is also controlled, at least to some degree by the driver would be the best cannon option. i know this would be bad for programmming.

Keep in mind also that unlike the CMU camera, this camera is a true ethernet camera with a built in web server. I absolutely agree that an auto tracking turret would be best, and I also agree that it would be beneficial to have a driver controlled turret either for teams that are light on programming or in some cases where a manual override is necessary (maybe you want to shoot to the left or right of the light post).

This year's competition makes it easier for manual control since it would be possible to have a laptop deliver realtime video from the camera and have a co-pilot manually rotate the camera into position. The co-pilot would have a robot view of what he/she is shooting at.

[=Martin=Taylor=]

Ok. I have to ask. I, too, have looked into using a vacuum to increase traction but I have run across a seemingly-insurmountable roadblock to the plot and it involves <R06>.

I had envisioned a tube or suction cup of sorts that would need to be pointed at the floor with a rubber shroud going around the edges that would just barely clear the ground. Here are my concerns though: 1) How do I make sure the shroud doesn't touch the ground, thereby violating <R06>, and 2) Will this even work to a good extent? Will there be too great of a loss of vacuum?

Additionally, I envision that the fan in the vacuum generator would suck air from under the robot and blow it out above it. Am I right in making this assumption?

Thanks!
-Matt

I believe the reason why 1771 was able to create so much force was becuase they had an excellent seal on the ball.

F = P x A

So if you can gradually build up pressure over a large area you can create a lot of force.

But without a seal on the floor, it'll be pretty darn near impossible to build up pressure.

[Dominicano0519]

Ok.. you got me.. Sorry a lil tired, bear with me here. Not sure if it's legal to gather game pieces for the purpose of reloading your dumper or shooter during the game after you dump or shoot your 7 that you can have on board.

of course its legal there is no rule that limits how many moon rocks you can hold at once; only empty cells

[Dominicano0519]

It all depends upon the driver. IF that driver can successfully figure out how to drive your robot the way you want it on the slippery surface, then it doesn't really matter what launching/dumping device you use.


This competition most relies on:
1) a great driver with lots of practice and experience
2) someone on a basketball team
3) a good strategy involving your two alliance members.

my .02

in AIM HIGH we had a hopper on our bot and we had tryouts for human players and the kids that were good at basketball where really bad at shooting little balls over a big wall the best two were actually short kids. dont know if it was a coincidence or anything

[Geeves]

After a lot of debate on our team, we seem to have settled on a turret launcher design. This is because a low powered launch can be used almost like a dump, so this gives you more scoring options, and a turret can be used to score anywhere, while a dump will most likely not have any sort of variety in scoring positions.

However, with this floor who knows what will happen.

[JesseK]

The interesting part comes with shooting. Both you and your target are (perpetually?) moving. to account for this, your software will need to be able to do motion prediction (probably using second-degree extrapolation of some sort), and using empirically or mathematically determined data, will be able to lock onto where the target will be. With a system like this in place, even a fast moving and evasive target will not be able to outsmart your turret... well unless its out of range or they move in a manner that second-degree extrapolation cannot predict.

Rather than having software choose how far ahead to lead the shot for the moving object, you could also simplify it at first so that the driver leads the shot right or left. Example: the camera finds the target, and the driver sees the angle at which the target is moving -- right to left across the front of the driver's robot. So the driver then presses one button to add a 3-4 degree offset to what the camera sees, such that the turret rotates to where the target is going. The camera still tracks the target and moves the turret as appropriate, but with the button down the software moves the turret to the target's bearing +/- the offset.

Defense corporations (such as Lockheed, the company I work for) spend millions of dollars on target motion tracking algorithms for precision targetting. This year's game doesn't require precision targetting (have you tried throwing the ball 27ft with any sort of accuracy?) so in the beginning I would plan on having a very simple way to adjust for moving targets. After all, if an opponent is moving at 10 feet per second they will more than likely crash into something, at which point they are no longer moving and are an easy target :cool:

Just my 0.02 though.

[martin417]

I believe the reason why 1771 was able to create so much force was becuase they had an excellent seal on the ball.

F = P x A

So if you can gradually build up pressure over a large area you can create a lot of force.

But without a seal on the floor, it'll be pretty darn near impossible to build up pressure.

With a gap, F still = P x A, it is just much harder to get P very big. last year we were able to achieve a delta P of ~1 PSI, so with a 24" seal diameter, we had ~450 lb.s of force. With a 1/8" gap around a similar sized circle, you have a leak area of around 10 in^2 if you look at this as on orifice, it would equate to a diameter of ~3 1/2". if you can move enough air to get a pressure drop of 1/2 PSI, you have effectively added 225 lb.s to your bot. (not to mention the added benefit of the Bernoulli effect at the rim, very cool!)

You also now have 225 lbs of force trying to lift a .90" thick fiber panel from the floor. (have you ever seen an animation of "storm surge" in the eye of a hurricane?) I cannot imagine the rules committee allowing this.

[2241]

My $.02? You will see almost no successful launchers this year. A few teams will pull it off, and we will marvel at them and slap our forheads and say OF COURSE, but most of the most successful teams will find a way to get balls reliably and quickly to the human players (including all four empty cells) and then have a human player who is absolutely golden scoring goals.

Someone namedropped Aim High; I'm thinking more like 2004 here.



i agree with this guy. there will be no successful launchers, however that doesn't mean there will not be launcher like items used, however i don't agree that the human players will have to be the reliable scorers. it will come down to dumping, if you can get a cage to dump with enough force, it will create a shotgun effect, and will score more effectively than anything else.

[AlexD744]

I think with the openings in the balls there would be very little accuracy from a shooter especialy from a moving target. There air passing through the ball would veer it in a different direction every time.

[JesseK]

I think with the openings in the balls there would be very little accuracy from a shooter especialy from a moving target. There air passing through the ball would veer it in a different direction every time.

We prototyped a shooter yesterday, and were able to get it to go 6 feet consistently within a 20-degree or so spread. Now all we have to do is adjust the speed, angle, and have a decent feeding mechanism. We should have an accurate shooter out to 10 feet or so...but 10-15 foot shots are our strategy anyways. It's possible, even with how squishy the balls are -- I'll post a picture sometime next week.

[CraigHickman]

i agree with this guy. there will be no successful launchers, however that doesn't mean there will not be launcher like items used, however i don't agree that the human players will have to be the reliable scorers. it will come down to dumping, if you can get a cage to dump with enough force, it will create a shotgun effect, and will score more effectively than anything else.

This is a VERY bold claim, especially for someone from such a high team number. I expect we'll see some teams with frighteningly accurate shooting mechanisms. After all, this is FIRST... and we've seen some pretty amazing systems in the past, many that no one thought could be done.

[Jon Stratis]

i agree with this guy. there will be no successful launchers, however that doesn't mean there will not be launcher like items used, however i don't agree that the human players will have to be the reliable scorers. it will come down to dumping, if you can get a cage to dump with enough force, it will create a shotgun effect, and will score more effectively than anything else.

the concept of shooting here is pretty similar to other situations where industrial solutions have been created. This of a battle ship on rough seas - the gun turrets mounted on the deck are able to compensate for the swells, keeping a constant trajectory lock on their target, even when the ship rolls 30 degrees from one side to the other. Likewise, the targeting system on a jet doesn't just tell you to squeeze your guns when you're pointed right at the guy - it compensates to lead the enemy plane with an intercept vector. A bomber doesn't drop when it's immediately over the target - it plots an intercept vector for the bomb.

If you look closely, you can find tons of examples where similar targeting/shooting/launching systems have been created and work - it's not unreasonable to expect that some teams will be able to achieve this as well. Especially since this technology has been available in military systems for decades.

[eschanz]

but you'll be moving around and the lag between the camera and shooting will through off the entire thing

[XXShadowXX]

the concept of shooting here is pretty similar to other situations where industrial solutions have been created. This of a battle ship on rough seas - the gun turrets mounted on the deck are able to compensate for the swells, keeping a constant trajectory lock on their target, even when the ship rolls 30 degrees from one side to the other. Likewise, the targeting system on a jet doesn't just tell you to squeeze your guns when you're pointed right at the guy - it compensates to lead the enemy plane with an intercept vector. A bomber doesn't drop when it's immediately over the target - it plots an intercept vector for the bomb.

If you look closely, you can find tons of examples where similar targeting/shooting/launching systems have been created and work - it's not unreasonable to expect that some teams will be able to achieve this as well. Especially since this technology has been available in military systems for decades.

This a great point. I would like to remind people that military could hit targets 23 miles away with there battleships in the 1940's using vacuum tubes, and punch cards we have 400Mhz processor, and a color camera...


but you'll be moving around and the lag between the camera and shooting will through off the entire thing


So make your math to account for it.

[jgraber]

With a gap, F still = P x A, it is just much harder to get P very big. last year we were able to achieve a delta P of ~1 PSI, so with a 24" seal diameter, we had ~450 lb.s of force. With a 1/8" gap around a similar sized circle, you have a leak area of around 10 in^2 if you look at this as on orifice, it would equate to a diameter of ~3 1/2". if you can move enough air to get a pressure drop of 1/2 PSI, you have effectively added 225 lb.s to your bot. (not to mention the added benefit of the Bernoulli effect at the rim, very cool!)

You also now have 225 lbs of force trying to lift a .90" thick fiber panel from the floor. (have you ever seen an animation of "storm surge" in the eye of a hurricane?) I cannot imagine the rules committee allowing this.

Thanks for the calculation examples.
On a "storm surge", the ocean surface is fluid. The Regolith is a solid.
If there were an airsource under the regolith, it might be a problem.
If the a cardboard box lists 40psi bursting strength, I don't think Regolith will have a problem. I'm looking forward to this being asked in the Q&A though.
I'm pleased to see that nearly every cool idea of mine has already been discussed in detail here already.

[martin417]

Thanks for the calculation examples.
On a "storm surge", the ocean surface is fluid. The Regolith is a solid.
If there were an airsource under the regolith, it might be a problem.
If the a cardboard box lists 40psi bursting strength, I don't think Regolith will have a problem. I'm looking forward to this being asked in the Q&A though.
I'm pleased to see that nearly every cool idea of mine has already been discussed in detail here already.

Regolith can NOT be treated as a solid. For purposes of discussion, think of it as a thick sheet of paper. A .090" Thick piece of fiber board is very flexible, and can deform a great deal (not as much as water obviously) It will be sitting on carpet, which is a porous medium, full of air at 14.7 PSI (in ideal terms at sea level, STP blah, blah, blah). If you expose the top side of this flexible member to a reduced pressure, say 14.2 (1/2PSI delta) you will have that 225 pounds of force pushing up on the sheet. One of the kids almost broke a window at the school last year by sticking the funnel against it with the vacuum on. If a quick thinking person hadn't turned off the vacuum immediately, the glass would likely have shattered. As it was, a witness on the other side of the glass said that it bowed inward a significant distance before the vacuum was shut off.

No cardboard box can withstand 40 PSI internal (or external) pressure (unless it is very tiny). Perhaps 40 PSI is the breaking strength of the cardboard? At my last job we had a cube shaped hydraulic tank pressurized to 5 (five) PSI. It was large, maybe 28" on a side. In order to prevent bowing of the sides, it was 3/8" thick steel with internal ribbing.

[The Pre]

if any of you were at the wisconsin regional in 2006 or saw team 111 (wildstang) play at nationals you know what a ball cannon can do...

or if any of you saw the team that won the chairmans award in 2006, team 111 wildstang... this was their ball cannon

i wouldn't be suppirsed if their 20 motorola engineer mentors and the students cranked out some ball launcher with teeth to go into the orbit balls and throw them... or even an arm to throw it.

in 2006 wildstang during the autonomous mode locked onto the target would dump their entire 5-10ball hopper strait into the top target. and when i say strait i mean a solid stream of 10 balls or so dumped in 5 maybe 6 seconds so i wouldn't doubt if there were some more attempts at copying this robot this year

[jgraber]

This game was practically designed to be a shooter game. ...
The basic logic i have hammered out is you would have a camera mounted on a pan/tilt on top of your turret. ... The camera is on the constant hunt for beacons that your team wants to shoot at. when it locates a beacon, there is a control loop that tells the camera to lock its pan/tilt such that the beacon is in the center(ish) of the feild of view.


Or use a much simpler (lower res) non-linear vision system.
Point the camera straight up at a cone or spherical reflector to get 360 degree vision.

[The Pre]

i think the 360 vision would be pritty crazy!!! props if you can get that to work, i've never used the camera before but that seems really confusing cause there are so many images that can be seen in 360 degrees, how will it know which one to go after?

[MrForbes]

This a great point. I would like to remind people that military could hit targets 23 miles away with there battleships in the 1940's using vacuum tubes, and punch cards

uhhh....I think that was done with mechanical computers and look up tables, since punch cards were used mostly for inventory control and the first vacuum tube computer was way way too top secret to let it go out in the field.

[kaszeta]

After evaluating the ball, I'm pretty sure the pinch-roller mechanism that our team has used before is still viable, but there's definitely a lot less accuracy than in Aim High. I'll probably help the students prototype something if I can get over to their workshop this week.

http://farm1.static.flickr.com/49/133042491_65c39db8f0.jpg (http://www.flickr.com/photos/kaszeta/133042491/)

[Daniel_LaFleur]

This is a VERY bold claim, especially for someone from such a high team number. I expect we'll see some teams with frighteningly accurate shooting mechanisms. After all, this is FIRST... and we've seen some pretty amazing systems in the past, many that no one thought could be done.

While I believe the number of highly accurate, reliable, and efficient cannons will be few and far between, I believe there will be a number of very successful designs.

[rvo1000]

As to the launching idea...we have been brainstorming on that and on a way of somehow accurately dumping the balls into other trailers as efficiently as possible, but no way so far. The dimensions are horrible to get around, but as we keep moving forward things are starting to work better.

[Ragnarokae]

I've seen a lot of confusion regarding the humans throwing the balls in.

<G09> TEAM Starting Positions – The PAYLOAD SPECIALIST is stationed in either of the FUELING STATIONS belonging to the ALLIANCE, or seated in the OUTPOST belonging to the ALLIANCE. The seat belt in each OUTPOST must be snugly fastened around the PAYLOAD SPECIALIST seated in this location. The determination of where the PAYLOAD SPECIALIST is stationed among these three possible locations is negotiated within the ALLIANCE before the start of the MATCH.

My understanding: Each alliance has a Payload Specialist in each fueling station, and one in the outpost (who is seated).

<G20> Handling MOON ROCKS – MOON ROCKS can be SCORED by ROBOTS or PAYLOAD SPECIALISTS. PAYLOAD SPECIALISTS can enter MOON ROCKS into play by launching them over the Alliance Station Wall or over/through the OUTPOST shield. MOON ROCKS can not enter the CRATER by being thrown around the end of the Alliance Station Wall. A violation will cause a PENALTY to be assigned.

My understanding: Only the Payload Specialist can throw.
Seems hard (to me) to throw a 9'' ball over a wall and land inside a 10.5'' target.... Maybe the Payload Specialist is meant to just get balls into the arena and get a very lucky shot.

<G39> TEAM Positions During TELEOPERATED PERIOD - During the TELEOPERATED PERIOD, the PILOTS and the COMMANDER may travel anywhere within the ALLIANCE ZONE (note that the ALLIANCE ZONE includes the local FUELING STATION). The PAYLOAD SPECIALIST must remain within the FUELING STATION to which they are assigned for the
entire MATCH. The PAYLOAD SPECIALIST stationed in the OUTPOST must remain in the OUTPOST seat during the entire MATCH (note that if the PAYLOAD SPECIALIST in the OUTPOST unbuckles the seat belt or stands up in this location, it may be considered a violation of <S01>). Each violation (stepping outside the designated area, or stepping across the PLAYERS LINE) will result in a PENALTY. Exceptions will be made in cases involving TEAM member safety.

My understanding: Payload Specialists can't move around to get a better shot.


I think that the focus needs to be on the robot getting the ball in, not on the human.

Anyone read the rules differently than I did?

[lingomaniac88]

Keep in mind that the game pieces can bounce off the poles on the trailer (both the ones around the perimeter and the one supporting the vision target). I think the payload specialists can score a lot of points if they're skilled enough.

[RossC]

I really like the idea of a turret. Being a second year team we decided to go simple and we're currently working on a basket design. I dont really like it because it was mostly done by a parent/mentor. It could also me by ego ;) A camera guided turret has a lot of advantages. I was wondering, why does the turret have to be refilled from the top? I was thinking that a vertical lift could be built to capture balls and lift them to a point. The turret could be suspended on the lazy susan? And powered by a motor. It wouldnt allow 360 movement but you could conceivably get 270. This would allow for 180 directly in front and straight back in order to ballance the score if necessary. I dont really know if this is a good idea but I wonder what others think. If things go drastically wrong we might reconsider.

[RyanCahoon]

I had envisioned a tube or suction cup of sorts that would need to be pointed at the floor with a rubber shroud going around the edges that would just barely clear the ground. Here are my concerns though: 1) How do I make sure the shroud doesn't touch the ground, thereby violating <R06>, and 2) Will this even work to a good extent? Will there be too great of a loss of vacuum?

Thought I'd contribute something back to the general knowledge pool:

My team prototyped a "ground sucker" today by attaching not 1, but 3 shop vac's to a rubbermaid storage container, inverted so the open side is pressed against the floor. When the container was flush against the floor, the vacuums had no problem completely collapsing it. At noticeably less than 1/16" above the floor, we felt some force, but anything above this there was no significant force. Not to say that this isn't possible, but as was pointed out earlier, 1771's success last year was largely due to getting a good seal on the ball. Our findings suggest that for this to work this year, a team would have to manufacture with very close tolerances, and any kind of flexing in the frame would cause the vacuum chamber to touch the floor; as such a robot such as this might not make it past inspection.

Has anyone else found different results?

--Ryan

[Ryan Albright]

I been reading this thread and drawing up a few designs, I still think a cannon is not going to be accurate enough.

I been looking at drawing a cannon design were it can hold at least 6 moon rocks. So after you have that loaded you have to hunt down a moving robot to score your points

I think the safest bet is to have a robot that can plow your balls to the human player so he can then shoot to score. This way the human player can adapt to how the other team is playing. I just have a feeling that a human player will be able to get more shots off then a cannon can.

I also think a hopper robot would be great, but then you are filling your hopper and hoping when you go to unload it that you can get them all in at one shot and if you miss, you just wasted all that time

I am sure I will see a couple teams design an awesome cannon that will make me eat my words. I do have to say my favorite robots each year are the ones that makes the hardest task look easy.

[XXShadowXX]

has anyone consider if the ball's recoil cause the robot to move?

[wkaeboarderNH]

our team just made a trailer to practice shooting. And its really isn't that hard, I'm terrible at basket ball but i made a good 80% on a stationary trailer. maybe a 60% on a moving one. not as hard as you may think. But the outpost with the seat may be difficult due to the hole in the wall.

[Dillon Carey]

How far away were you shooting from? were you shooting over the alliance station wall? did you count the robot being bigger then the trailer and getting in the way? and did you count the 5 other robots on the field to get in the way?

[wkaeboarderNH]

we had our mentor who is about 6 foot 5 stand with a noodle and we shot over the noodle. But no we didn't account of the robot being taller than the trailer. and not with 5 other robots. i think we were shooting from about 15 - 20 feet away.

[Robert Cawthon]

[QUOTE=Petey;792201]My $.02? You will see almost no successful launchers this year. A few teams will pull it off, and we will marvel at them and slap our forheads and say OF COURSE, ...QUOTE]

After three years of helping design robots, I am a charter member of the "Flat Forehead Disease" club, where I see how some other team has done something, slap my forhead, and say "Why didn't I think of that!" One things for sure. Someone will come up with a neat, simple way to shoot the balls. Just hope it will be my team. Good luck, everyone. Enjoy the build.:D

[Robert Cawthon]

This a great point. I would like to remind people that military could hit targets 23 miles away with there battleships in the 1940's using vacuum tubes, and punch cards we have 400Mhz processor, and a color camera...



So make your math to account for it.

And six weeks.

[shannieliz]

in my opinion, i think not to do it.
i mean, if you throw the ball, and it goes over the wall, it counts as a penalty.
so if you throw the ball at someone's trailer, and they are near the wall, and they move, it could go over the wall and then you would get a penalty.

[EricH]

in my opinion, i think not to do it.
i mean, if you throw the ball, and it goes over the wall, it counts as a penalty.
so if you throw the ball at someone's trailer, and they are near the wall, and they move, it could go over the wall and then you would get a penalty.Not necessarily. I see no rule against throwing things off the field. HOWEVER, under <S01>, you could be dealt with for doing it too often.

Also, the balls will be returned to the field in the appropriate place.

[DM7000]

Not necessarily. I see no rule against throwing things off the field. HOWEVER, under <S01>, you could be dealt with for doing it too often.

Also, the balls will be returned to the field in the appropriate place.

I think you might get a penalty if you are purposely throwing them off the field. As for <S01>, depending on how fast those balls are going and how often I don't think it posses a safety hazard.

[EricH]

I think you might get a penalty if you are purposely throwing them off the field. As for <S01>, depending on how fast those balls are going and how often I don't think it posses a safety hazard.
Quote a rule.

As for whether or not they are a safety hazard, one or two by accident probably won't be. Several dozen by "accident" would be.

[DM7000]

Quote a rule.

As for whether or not they are a safety hazard, one or two by accident probably won't be. Several dozen by "accident" would be.

I could not find a quote to back my statement up. Yet if you're not purposely throwing them off the field they would not pose a safety hazard.

[EricH]

I could not find a quote to back my statement up. Yet if you're not purposely throwing them off the field they would not pose a safety hazard.That remains to be seen, and must be judged individually.

[BlueMyrian]

Our team decided to build a shooter robot and were looking to use the ball pitcher/flywheel method to shoot the balls only to find that after repeated shooting the balls would break. We now have a prototype shooter that shoots quickly and fairly accurately with a range near 15 feet.

My question is if anyone else has encountered a problem with their balls breaking under the pressure of a ball pitcher? Could a ball pitcher be built to be both powerful and accurate without putting to great a stress on the Orbit balls?

As far as the problems with trying to load a turreted shooter, we are currently brainstorming about how to build a hopper that will effectively hold a number of balls and still have our shooter as high as possible on our robot. Since our shooter needs to be top loaded, we're working on a way to fit the shooter between 40'' and 50'' on our robot and it becoming obvious that it can be done with a little creative designing.

[GaryVoshol]

Was anyone penalized last year for shooting a Trackball off the field? (I know, obligatory reference to Dave, "This year's rules are not the same as last year's rules." But they are - there is no reference in 2008 or 2009 to a penalty for shooting a ball off the field.)

And back on topic, how many posts did we see last year predicting the inability of flinging (by whatever method) a Trackball over the Overpass?

[XXShadowXX]

seams very possible to me, DIFFICULT, but all to possible. Somebody will pull it off and my team should work our butts off to.

My current design, would fire balls around 85 feet at 420 balls per minute (7 balls per second), showed it to my coach and he said that were trying to score, not kill the other alliance. The only difficulty I'm having is working on a leading algorithm, why can't calculus be easy.

[wcamp1742]

My team prototyped a design using a similar flywheel mechanism to Aim High's design and the wheels caused so much friction it just ripped the cover off the balls while only shooting the ball about 2 ft. I don't think we'll see many this year along those lines. For this reason we're moving to more of a catapult type launcher that can get about 5-8 shots off in autonomous.

[Luke Pike]

seams very possible to me, DIFFICULT, but all to possible. Somebody will pull it off and my team should work our butts off to.

My current design, would fire balls around 85 feet at 420 balls per minute (7 balls per second), showed it to my coach and he said that were trying to score, not kill the other alliance. The only difficulty I'm having is working on a leading algorithm, why can't calculus be easy.

Depending on how your shooter works, you could do it with just one-dimensional motion equations and some trig. Ignoring air-resistance, of course.

As for your 420 balls per minute, you better have a way to get a lot of balls fast.

[pfreivald]

My team prototyped a design using a similar flywheel mechanism to Aim High's design and the wheels caused so much friction it just ripped the cover off the balls while only shooting the ball about 2 ft. I don't think we'll see many this year along those lines. For this reason we're moving to more of a catapult type launcher that can get about 5-8 shots off in autonomous.

How far are you planning to throw the balls? We have been playing with a flywheel type design, and can get 8-10 feet in a decent arc without any damage to the moon rocks.

I'm dubious that any *long range* shooting would be particularly accurate anyway, what with the aerodynamics of the Orbit Balls being rather, um, unpredictable.

Patrick

[Jon Stratis]

FWIW, the 85 feet XXShadowXX claims he can do is a bit overkill - the field is only 54 feet long!

With the field being 27 feet wide at any given point, you really only need 14 feet on a shooter to be able to control half the field. The closer your robot is to your target, the more accurate you'll be - that much is almost a certainty. While 85 feet is definitely cool - and could come in handy with off season events and pep rallies - it not needed here at all.

[IKE]

We made a turreted shooter for Aim High in 2006. There is a lot of design detail in the Behind the Design Book. I highly suggest reading that entry. A couple of things you will not find in the book directly:

Ball sorting is extremely difficult! in 2006 most Hopper style designs jammed when trying to serialize a bin of balls.

We went for a single wheel to a double wheel system, because while the single wheel was accurate, the spin on the ball descored about 50% of the shots we made. Also Spin energy was significant portion of the launch energy equation, and that was with a "solid" ball. Since the moon rock is a bit more of a "hollow sphere", this will become an even larger portion of the equation.

Turret to conveyor interface is super critical. You need a feed mechanism that consistantly delivers balls to the cannon at the same speed and position in order to be accurate.

With over 1500 teams, someone will have an amazing cannon this year.

As for range: Shaddowxx please recheck your numbers. To deliver that many balls at that rate of speed will require a ton of power. Not saying it can't be done, but that will require a lot of the power motors, or a lot of stored energy.

[JesseK]

seams very possible to me, DIFFICULT, but all to possible. Somebody will pull it off and my team should work our butts off to.

My current design, would fire balls around 85 feet at 420 balls per minute (7 balls per second), showed it to my coach and he said that were trying to score, not kill the other alliance. The only difficulty I'm having is working on a leading algorithm, why can't calculus be easy.

Spray and pray? :confused:

[omgapirate]

Has your team considered actively dumping? (Dumping but with something propelling the balls out of the dumper)

[ZakuAce]

Spray and pray? :confused:

Awesome, I was going to say this.

Machine guns are great and all, but a sniper puts you in constant fear. I think being consistant will be crucial.

Slow and steady wins the race, right?

[IKE]

Awesome, I was going to say this.

Machine guns are great and all, but a sniper puts you in constant fear. I think being consistant will be crucial.

Slow and steady wins the race, right?

Sniper relies on the accuracy, and elements of surprise and disguise. In 2006 our ball sorter would begin spinning about 1.5 seconds before the first ball fired. It ended up being a giant "HIT ME" sign for our opponents. The gorrillas on the other hand had speed and accuracy, but limited payload. They could deliver about 10 shots in 1.5 seconds, but could really only deliver those 10 and would have to reload.

Just some more food for thought? I think we are going to make a ghillie suit for our robot made out of FRP, so teams won't see us sneak up on them...:D
http://www.theghilliesuits.com/
http://video.google.com/videosearch?hl=en&q=snow+ghillie+suit&um=1&ie=UTF-8&sa=X&oi=video_result_group&resnum=7&ct=title#

[kramarczyk]

Just some more food for thought? I think we are going to make a ghillie suit for our robot made out of FRP, so teams won't see us sneak up on them...:D
http://www.theghilliesuits.com/
http://video.google.com/videosearch?hl=en&q=snow+ghillie+suit&um=1&ie=UTF-8&sa=X&oi=video_result_group&resnum=7&ct=title#

And those that try to hit you will just slide right off. :p

[JKWarrior]

erm.... in my opinion i think launching is out of the question.... better just to throw from the stations and to sort of dump the balls into the other alliances trailer

I agree, pinning and dumping is more accurate:D

[Jon Stratis]

I agree, pinning and dumping is more accurate:D

it may be more accurate, but the question is how easy/quick is it? For that strategy, you're assuming you can first catch up to another robot, second pin them against the wall, and third dump in them before they have a chance to get away (keep in mind that they will have more traction with the carpet border). You also have to consider how much time all of that will take - time a shooting bot will be collecting and firing balls non-stop.

There are trade-offs for both types of robots with the game design. I don't believe there is one type (shooting or dumping) that can be considered "better" overall than the other. At least, not at this point.

[Hanna2325]

I see a canon as a good option, at least compared to a dumper; just because the height a bucket would have to reach for dumping along with necessary size and good stalking skills, make a dumper seem a lot riskier and less likely (according to our prototypes)...but who knows, there still are a ton of complications with the canon...:yikes:

[pfreivald]

So, a flywheel shooter is spectacular only in its mediocrity. It seems that most of your energy goes into deformation of the ball, and not into kinetic energy. Anybody having any success at throwing the ball any reasonable distances?

Patrick

[acdcfan259]

So, a flywheel shooter is spectacular only in its mediocrity. It seems that most of your energy goes into deformation of the ball, and not into kinetic energy. Anybody having any success at throwing the ball any reasonable distances?

Patrick

We're prototyping a piston launch system. We can launch the balls about 10 feet and they can make it into the trailer.

[pfreivald]

We're prototyping a piston launch system. We can launch the balls about 10 feet and they can make it into the trailer.

'About 10 feet...' How do you control the distance you launch? A variable-force piston?!?

Patrick

[acdcfan259]

'About 10 feet...' How do you control the distance you launch? A variable-force piston?!?

Patrick

Key word, prototyping.

[gorrilla]

Key word, prototyping.


words to live by......

were working on combining our harvester with our "vomiter"....so much surgical tubing so little time......

[Hanna2325]

We have a good prototype/start up now, a video (of one of our not so hot rounds) is here http://www.youtube.com/watch?v=S04ub3plTSY

[TheOtherGuy]

We finally got our prototype working up to our standards, here's a video:

http://www.youtube.com/watch?v=ZIlRWFW9jsk *

We opted for the backspin and shorter distance... first time we've made a shooter and we're enjoying the consistency when firing multiple shots in a row :) something like 3 balls/second.

*No freshmen were harmed in the making of this film.

[MrForbes]

Now, if we could only to get it to work better than our standards! :)

[Jon Stratis]

Yeah, that prototype definitely proved it's possible - we could hit the trailer 16 feet away! And on top of that, what the team learned from the prototype should allow us to improve upon it and increase range and accuracy.

[Hanna2325]

We finally got our prototype working up to our standards, here's a video:

http://www.youtube.com/watch?v=ZIlRWFW9jsk *

We opted for the backspin and shorter distance... first time we've made a shooter and we're enjoying the consistency when firing multiple shots in a row :) something like 3 balls/second.

*No freshmen were harmed in the making of this film.

Good Job! and thats just the prototype :D

[Tom Line]

I think the discussion between dumping and shooting should be heavily skewed towards shooting.

After all, a well-made shooter should be able to "shoot" balls in VERY quickly from a foot or two away, nearly rivalling a "dumper".

[IKE]

Uh oh:yikes: beware of flying Pi's.

[Doug Leppard]

This is a good discussion, thanks for the videos.

[=Martin=Taylor=]

We finally got our prototype working up to our standards, here's a video:

http://www.youtube.com/watch?v=ZIlRWFW9jsk *

We opted for the backspin and shorter distance... first time we've made a shooter and we're enjoying the consistency when firing multiple shots in a row :) something like 3 balls/second.

*No freshmen were harmed in the making of this film.

Great work guys! I'm always impressed with what you come up with.

I'm not sure I understand the need to shoot 3~4+ feet. I mean, its not like in '06 where we couldn't get right up to the goal. Why not just drive right up and squirt them in?

If you think you can score from 8~10+ feet away why not just use the HP's? They'll have better accuracy than the bot.

[Jon Stratis]

Great work guys! I'm always impressed with what you come up with.

I'm not sure I understand the need to shoot 3~4+ feet. I mean, its not like in '06 where we couldn't get right up to the goal. Why not just drive right up and squirt them in?

If you think you can score from 8~10+ feet away why not just use the HP's? They'll have better accuracy than the bot.

Mostly because of the Regolith. With the new playing surface and wheels, robots won't be nearly as maneuverable. Additionally, it'll take more time to get up to speed than it has in past years, and turning will be much more difficult - especially with a trailer. When you add all that up, chasing after another robot to dump (or short range squirt) can become challenging. They'll hit the wall first, and have added traction from the carpet to get going in another direction before you can get in range. Plus, all the time spent chasing someone down is time you spend with a full hopper of balls, unable to collect and shoot more. With a longer range cannon, you're much more likely to be able to constantly fire balls, and just worry about gathering them and staying away from other players.

All that being said, our prototype shows the potential range. If the other alliance's robots are evenly distributed across the field, then there will be one within 10 feet of us a vast majority of the time.

[MrForbes]

I'm not sure I understand the need to shoot 3~4+ feet. I mean, its not like in '06 where we couldn't get right up to the goal. Why not just drive right up and squirt them in?

We expect them to see us coming, and run away as fast as they can!

(always the optimist)

And we do expect to get a lot of scoring help from the PSs, but it is also going to be a bit of a challenge for them to score when the goals are over 8-10 feet away. Every bit helps!

[sdcantrell56]

I think there is no way that a human player can outscore a well designed turreted shooter. Think about it. A shooter can keep a constant velocity and launch angle unlike a human. Also a shooter is more consistent in calculating distance. Finally, a shooter can fire balls much more rapidly. We are aiming for ~3 balls per second.

Squirrel how much compression are you guys running. We were doing 2" and it was working very well

[s_forbes]

Squirrel how much compression are you guys running. We were doing 2" and it was working very well

We've got about 1" compression at maximum. The distance is set up so that when the ball is being contacted on the holes, it just touches. If you haven't yet, try experimenting with tread on the support side of the turret in addition to the wheel...

[Dan2081]

2081s concept will be a combination between dumping and shooting. It will be able to shoot medium to short distances, however we are going to make the angle of trajectory changeable so that when we get near an opponent, we can shoot the balls at a slower speed, still unloading them quickly and aiming them horizontally, if not downwards

[katyrobo2177]

Our team is going after a somewhat similar idea with a cannon.
Does anyone have any idea what the angle of the camera image is?

[pfreivald]

2081s concept will be a combination between dumping and shooting. It will be able to shoot medium to short distances, however we are going to make the angle of trajectory changeable so that when we get near an opponent, we can shoot the balls at a slower speed, still unloading them quickly and aiming them horizontally, if not downwards

Isn't a more elegant (and mechanically simpler) solution for range to change the speed of your shooter, instead of the angle of the shot?

Patrick

[EricH]

Isn't a more elegant (and mechanically simpler) solution for range to change the speed of your shooter, instead of the angle of the shot?

Patrick
It can be very mechanically simple to change the angle. A number of teams in the 2006 Behind the Design book (and some that didn't make it) changed their shooting angle to adjust for their position relative to the target, in ways that made the rest of us say, "Why didn't I think of that?"

[pfreivald]

It can be very mechanically simple to change the angle. A number of teams in the 2006 Behind the Design book (and some that didn't make it) changed their shooting angle to adjust for their position relative to the target, in ways that made the rest of us say, "Why didn't I think of that?"

Sure thing -- the big winners locally in Aim High did one or the other, too.

I just noted the large team number and figured there was probably some inexperience there, so I thought I'd suggest another possibility. I think either one could work well, depending upon implementation.

Patrick

[JKWarrior]

you could make the the balls feed from the back, have a large opening and have the lazy susan(or turntable) on top. Useing a shaft run up from the plate attached to the turntable and put a sproket on the shaft connected to a motor.

[gorillamonky]

My $.02? You will see almost no successful launchers this year. A few teams will pull it off, and we will marvel at them and slap our forheads and say OF COURSE, but most of the most successful teams will find a way to get balls reliably and quickly to the human players (including all four empty cells) and then have a human player who is absolutely golden scoring goals.



not at all, there is absolutely no way for a human shooter to be accurate enough to design a bot around them. if you have done any testing at all you would have realized this, unless you have some crazy good shot on your team

[Daniel_LaFleur]

not at all, there is absolutely no way for a human shooter to be accurate enough to design a bot around them. if you have done any testing at all you would have realized this, unless you have some crazy good shot on your team

And you believe that a robot will shoot better?

TBH, There will be robots that shoot well, and robots that shoot poorly ... and Humans that shoot well, and humans that shoot poorly. In the end, it'll be fun to watch :D

[gorillamonky]

And you believe that a robot will shoot better? :D

no, i am reasonably sure that a robot will shoot even worse than a human, but i have no testing to prove this. my team is going with a dump type

[DuzieInuKenshin]

We've been looking at both designs and seeing what we can get done on both sides of the equation. For the launcher we figured out how to get the camera to track colors and have a sketched out equation to calculate range based on trigonometry, however we've run into an issue on how to launch the balls. Our main design was to spin two wheels in opposite directions and accelerate the ball through them, but we can't get enough surface area between the ball and the wheels to really launch them. There is also the fact that it would be hard to get much capacity to hold balls on the robot. We figure 5 balls on the robot at one time is a stretch. The plus of a shooter would be the ability to better choose targets.

Our dumping design has just as many issues. As was mentioned earlier, really efficient dumping would mean essentially stopping both robots. Although we did find that a dumper is between 50 and 75% accurate within a foot, so if you could design a dumper that would release balls quickly you could probably do a drive by dump. The capacity on a dumper is 6 to 7 balls. Using a dumper, though, you wouldn't be as able to choose your target, and once you dropped your balls you would have to collect an entire new set.

I think the key factors are going to be time and manueverability. There are 2.5 weeks left. At some point one design may be faster to build than another and it may be more beneficial to have the extra couple days to program and fine tune than to do the other design. And, on a frictionless surface, you're going to want as much control over your bot as you can get.

[KenGriffin26]

This is our launcher prototype. It has a bit of a lexan "nose" that fits over it, and the aluminum version will have it completely one piece, pivots and raises.

http://www.youtube.com/watch?v=Dxqu9wyhPu4&feature=channel

[Vikesrock]

We figure 5 balls on the robot at one time is a stretch.

The capacity on a dumper is 6 to 7 balls.

This depends on design.

Our shooter has the capacity to hold up to 11 balls as it is currently designed and I expect that some teams will be capable of holding even more than this.

[TheOtherGuy]

Just as an update, we upped the speed of our cannon to 3700 rpm, and the diameter of our roller is now 5" with about 1.5" compression. Here's the vid:

http://www.youtube.com/watch?v=qNH4hb7crFU

it's very consistent, we'll be testing distance controllability with varying the motor speed tomorrow (just got electronics board mounted). Gets ~10 ft and probably still around 3-4 balls per second (also need to test spin-up time).

[AlexD744]

Our team has veered from using a shooter because we figure the balls will veer because of their holes. However, I have seen designs that seem to be quite accurate. Therefore I don't know.

[eschanz]

Heck yea, we're doing a shooter too. I prototyped a successful launcher yesterday that went 8 feet, and the students are almost done prototyping a shooter today. Combine it with a turret and camera and you're golden -- they key is the type of shooter. Get creative, you don't have to use flywheel style...

We're not using a hopper though. Our overall game strategy dictates that we won't have the time to collect large amounts of moonrocks, so we'll fire what we get almost as soon as we get it. We'll have a max of 6 moonrocks in our bot at once, and probably then only at the start. This is for simplicity's sake though...

thats sweet!!