How to pick up a frisbee

[kuraikou]

My team (4183) has been having trouble deciding if we want to make our robot able to pick up Frisbees and if we want to how can we easily pick them up and deposit them in our shooter.

any ideas?

[Andrew Lawrence]

That plunger and racket idea looked pretty good to me....



On a more serious note, one idea we had was using a dustpan-like mechanism to get under the frisbees and flip them into our hopper.

[Twest3259]

the simplest solution i could think of was to pick them up like you'd pick up a basketball. just use rollers to suck it up.

[kuraikou]

the simplest solution i could think of was to pick them up like you'd pick up a basketball. just use rollers to suck it up.

But than how would you put them into your shooter?

[Twest3259]

But than how would you put them into your shooter?

a conveyor or elevator system that rolls them up and straight into the shooter or hopper.

[Wildcats1378]

That's the easy part. The hard part is determining which way they are facing, and then how to flip them.

I think that picking up the frisbees is going to be a waste of time. There's a pretty small number sitting on the field, and the main way to get most of the frisbee's is through the feeding station.

However, I was thinking of a conveyer system which would have a little hatch open. Above the hatch would be a suction cup on a piston. It would go down, pick up the frisbee, then deposit it on the conveyer belt. Simple and fast. Doesn't solve the upside down issue though...

[ENeyman]

One solution is simply to ignore the upside down discs and just go for the ones that are right side up.

On that note, however, does the manual explicitly state that the discs on the field at the beginning of the match are face up? I assume so, but I can't find a specific rule.

[EricH]

I think it's quite possible to build a launcher that will toss discs equally well when they're upside down as right side up, or near enough that there isn't a noticeable difference at the type of ranges you might see in this game. I've heard a human throw type called the "Hammer" or the "Tomahawk"; it goes from vertical to inverted in flight and stays inverted. It also travels a fair ways, if thrown correctly.

[Rishabhgadi]

I was wondering..Would it be completely legal to back up against the deposit slots and have them land in a hopper right away?

[MetalJacket]

yep

[GaryVoshol]

I was wondering..Would it be completely legal to back up against the deposit slots and have them land in a hopper right away?

Sure.

[gabrielau23]

I was wondering..Would it be completely legal to back up against the deposit slots and have them land in a hopper right away?

I don't see a problem with that....I was going to suggest that to my team, anyway....

[z_beeblebrox]

Perhaps have a toilet plunger end on a pneumatic arm come down and stick on the disc and then pull it back up. On the way up, the disc would hit a mechanical stop and fall of the plunger into a ramp or hopper.

[gabrielau23]

I'm not so sure a toilet plunger would work so well....for one thing, the suction isn't actually that great. The hard rubber pretty much prevents that. I think the best thing is to just use a dustpan-ish mechanicsmand drive around with that.

[Rishabhgadi]

You could in terms just build an arm that has a tray of sorts and just dumps 3 at a time?

[kuraikou]

Perhaps have a toilet plunger end on a pneumatic arm come down and stick on the disc and then pull it back up. On the way up, the disc would hit a mechanical stop and fall of the plunger into a ramp or hopper.

http://www.youtube.com/watch?v=t8tEbK8S7-8

a cool plunger pickup system

[Joon Park]

I think that picking up the frisbees is going to be a waste of time. There's a pretty small number sitting on the field, and the main way to get most of the frisbee's is through the feeding station.

Although I agree that we're going to see the feeding slots used more than ever in the past few years, picking up discs from the field will be a HUGE advantage, because it's actually something that not all the teams will be able to do, nevertheless do it well. There will be quite a few discs lying around the field from missed shots, I imagine.

I think it's quite possible to build a launcher that will toss discs equally well when they're upside down as right side up, or near enough that there isn't a noticeable difference at the type of ranges you might see in this game. I've heard a human throw type called the "Hammer" or the "Tomahawk"; it goes from vertical to inverted in flight and stays inverted. It also travels a fair ways, if thrown correctly.

I addressed this in another post, but if you're planning to throw the disc with any resemblance of accuracy over any distance greater than, say, 3ft, you're going to have to do it the way it was intended: right side up, and with spin. I guarantee you we will see 0 robots that emulate the human throw called the hammer. A hammer is hard for a human, it'd be even harder for a robot. (Source: ultimate player)

[Rishabhgadi]

Perhaps have a toilet plunger end on a pneumatic arm come down and stick on the disc and then pull it back up. On the way up, the disc would hit a mechanical stop and fall of the plunger into a ramp or hopper.

Or have an arm, so that you can extend and just release?

[ttldomination]

I addressed this in another post, but if you're planning to throw the disc with any resemblance of accuracy over any distance greater than, say, 3ft, you're going to have to do it the way it was intended: right side up, and with spin. I guarantee you we will see 0 robots that emulate the human throw called the hammer. A hammer is hard for a human, it'd be even harder for a robot. (Source: ultimate player)

http://www.youtube.com/watch?feature=player_detailpage&v=iyE4Ir6dkY8#t=76s

Seems to me that if you got it going good enough, an upside down discs does nearly as well.

- Sunny G.

[pfreivald]

I addressed this in another post, but if you're planning to throw the disc with any resemblance of accuracy over any distance greater than, say, 3ft, you're going to have to do it the way it was intended: right side up, and with spin. I guarantee you we will see 0 robots that emulate the human throw called the hammer. A hammer is hard for a human, it'd be even harder for a robot. (Source: ultimate player)

I'm going to go out on a limb here and say you're wrong (http://www.youtube.com/watch?v=iyE4Ir6dkY8) on this point.

[gabrielau23]

Although I agree that we're going to see the feeding slots used more than ever in the past few years, picking up discs from the field will be a HUGE advantage, because it's actually something that not all the teams will be able to do, nevertheless do it well. There will be quite a few discs lying around the field from missed shots, I imagine.



I addressed this in another post, but if you're planning to throw the disc with any resemblance of accuracy over any distance greater than, say, 3ft, you're going to have to do it the way it was intended: right side up, and with spin. I guarantee you we will see 0 robots that emulate the human throw called the hammer. A hammer is hard for a human, it'd be even harder for a robot. (Source: ultimate player)

Robot in 3 Days would like to disagree with you about the upside down part:
http://www.youtube.com/watch?v=iyE4Ir6dkY8

[Kellen Hill]

I'm not sure that picking up discs off the floor will be very important, but this is something I hope to see on at least a few robots.

http://www.diginfo.tv/v/11-0121-r-en.php

[AdamHeard]

I'm going to go out on a limb here and say you're wrong (http://www.youtube.com/watch?v=iyE4Ir6dkY8) on this point.

I don't see them making them in any sort of goal, and there is no indication of precision whatsoever.

Don't assume this video proves anything.

[Kpchem]

I'm not sure that picking up discs off the floor will be very important...

The one big advantage I see to floor loading discs is that if you miss a shot, you can pick it up again and fire immediately, instead of having to cross the field to reload. I feel that for some teams, this will prove to be a big factor in their performance, because they can score every disc around them.

[z_beeblebrox]

From the accuracy of prototypes and the large size of the goals, I don't think that good teams are going to be missing many shots. In addition, as I learned last year, just the driving to pick up game pieces requires great precision and can be time consuming. It seems easier to zip back to your driver station and quickly load 4 more from that hard-to-deplete supply.

[falconmaster]

I'm not sure that picking up discs off the floor will be very important, but this is something I hope to see on at least a few robots.

http://www.diginfo.tv/v/11-0121-r-en.php

That was just awesome! Thanks

[ksafin]

Yeah, our team initially planned to do a mechanism to pick up discs from the floor but..

Guys lets be honest.. each team on your alliance gets 17 discs. I think in the 1:30 tele-op (not including end game) it's very reasonable that you'd be able to traverse the field to your hopper, get filled, go back, score, and repeat to get pretty much just those 17 discs in (if even).

In other words, don't worry about picking up discs. Go get filled because it takes attention off of picking up and elevators and focuses on filling some hopper on your robot that'll go straight into a shooter. In 1:30, I can't see getting more than 17 discs across anyway. If you do, that's excellent, just be happy with that!

[McGurky]

To all teams reading this, make sure that you test any prototype on carpet similar to that being used on the official field.

A design that might work on your shop floor could very well fail on carpet.

In addition, I think teams should look at how many shots missed the hoops last year. Expect about that many disks to miss their target and end up on the ground.

[MrTechCenter]

Sure it's much easier to go across the field to your alliance's feeding station and grab four discs, but think about it. After autonomous mode, you'd have 3 robots fighting for discs at the alliance feeding station. Sure there's enough slots for all three to be filled at the same time, but realistically, all three would not be fed at the same time, as the corner is very small and can barely fit all three robots there assuming that they are average sized. Plus, you have to worry about d-bots getting in your way and defending your feeding station. If you can pick up discs from the ground quick enough, it would be a HUGE advantage to you, and it would be beneficial during autonomous too, because then you can pick up discs from the ground and shoot them during auton for additional points (after you've shot your preloaded discs). And as for there being a lack of discs on the ground, the goals are fairly large but with the amount of defense that's possible, I can see a LOT of missed shots when a d-bot knocks a scoring bot off of alignment. We saw this last year with fender-shooters and d-bots.

[Garten Haeska]

When I first saw the animation I thought that it would be hard to pick them up off the ground, then when I saw that we would potentially be suspended 90in. In the air, I quickly said, good shooter, good arm, no pick up. Yes if your alliance gets 17 frisbees each that means why pick up when you can just recieve. Just my honest opinion. Just remember this year, the lower the weight the easier the climb.

[gabrielau23]

Could a robot block the path to the feeder station by just sitting there?

[ttldomination]

I don't see them making them in any sort of goal, and there is no indication of precision whatsoever.

Don't assume this video proves anything.

Just eyeing those shots, they seem enough to the point of at least making the 2 pointers.

But, as is popular, "assuming makes an $@#$@#$@# out of you and me". At least for my team, this warrants a couple of runs through our prototypes.

- Sunny G.

[Garten Haeska]

I forget which rule, but your robot can't purposefully block an area of field and then they aren't allowed to touch your robot when they are at the feeder station.

[nathan_hui]

I addressed this in another post, but if you're planning to throw the disc with any resemblance of accuracy over any distance greater than, say, 3ft, you're going to have to do it the way it was intended: right side up, and with spin. I guarantee you we will see 0 robots that emulate the human throw called the hammer. A hammer is hard for a human, it'd be even harder for a robot. (Source: ultimate player)

One thing that I will warn, and this is something my fellow mentors at 2473 keep emphasizing, is that just because a human can or cannot do something does not mean that a robot can or cannot do it. There are many of examples where robots can do things humans cannot, and where robots cannot do things humans can. We and robots are two very different organisms - it is wrong to compare the two.

[MrTechCenter]

I forget which rule, but your robot can't purposefully block an area of field and then they aren't allowed to touch your robot when they are at the feeder station.

Yes, I also recall that rule. It states that a robot cannot intentionally block and prevent the "flow" of the match. But this rule has been in the manual for quite some time, as I remember. Like in Logomotion, bots patrolled the middle of the field and blocked bots from crossing. I don't remember penalties every being called against this. The only defensive penalties I've seen is contact in a protected zone (which would be the feeder station and when a bot is touching the pyramid in this case). So this leads me to believe, either the refs aren't calling these penalties, or this isn't considered blocking the "flow" of the match. And if that isn't blocking the flow, I don't know what would be considered so.

[nixiebunny]

I get the feeling that this is going to be the biggest argument of the build season.

I personally think that our second-year team is better off not trying to get them off the floor, and instead concentrating our efforts on building a feeder-fed bot in three weeks and practicing with it for the next three weeks.

That will allow us to be consistent scorers, so we won't be littering the field with frisbees. It will also get us ranked highly in the qual matches, which is where we want to be.

[rutzman]

Yes, I also recall that rule. It states that a robot cannot intentionally block and prevent the "flow" of the match. But this rule has been in the manual for quite some time, as I remember. Like in Logomotion, bots patrolled the middle of the field and blocked bots from crossing. I don't remember penalties every being called against this. The only defensive penalties I've seen is contact in a protected zone (which would be the feeder station and when a bot is touching the pyramid in this case). So this leads me to believe, either the refs aren't calling these penalties, or this isn't considered blocking the "flow" of the match. And if that isn't blocking the flow, I don't know what would be considered so.

G25) ROBOTS on the same ALLIANCE may not blockade the FIELD in an attempt to stop the flow of the MATCH. This rule has no effect on individual ROBOT-ROBOT interaction.
Violation: TECHNICAL FOUL

[MrTechCenter]

G25) ROBOTS on the same ALLIANCE may not blockade the FIELD in an attempt to stop the flow of the MATCH. This rule has no effect on individual ROBOT-ROBOT interaction.
Violation: TECHNICAL FOUL

So, it would be a penalty to, say, park in front of the opposing alliance's feeder station. But it's ok to prevent them from crossing the field to get to their feeder station. Is this correct?

[Randomness]

So, it would be a penalty to, say, park in front of the opposing alliance's feeder station. But it's ok to prevent them from crossing the field to get to their feeder station. Is this correct?

No - it is a penalty to have multiple robots block the entire opposing alliance from crossing. A single robot can play defense without any penalties, unless they touch an opposing robot who is touching either the pyramid or the loading area.

The key parts from

G25) ROBOTS on the same ALLIANCE may not blockade the FIELD in an attempt to stop the flow of the MATCH. This rule has no effect on individual ROBOT-ROBOT interaction

are "ROBOTS" plural and "no effect on individual ROBOT-ROBOT interaction."

[pfreivald]

I don't see them making them in any sort of goal, and there is no indication of precision whatsoever.

Don't assume this video proves anything.

Just eyeing those shots, they seem enough to the point of at least making the 2 pointers.

But, as is popular, "assuming makes an $@#$@#$@# out of you and me". At least for my team, this warrants a couple of runs through our prototypes.

- Sunny G.

Right. Using the pole and the edge of the canopy as guides, you can get a pretty good idea of the precision.

[pandamonium]

Tape... :-D

[donnie99]

Addressing two points in this thread:

1. A "Hammer" or "Tomahawk" is thrown exactly as the name would suggest, over the head. This is not the direction a disc would travel in a flat path. The purpose of this type of throw is to travel at a greater height than distance. This is why human players would use this to throw over the alliance wall. Robots could possibly throw this way. As every new year comes, some team somewhere will make it happen if they try hard enough.

2. The rules state that you may block people at any point as long as they are not touching their pyramid, their safe zone next to the feeder station, and you do not pin them for more than 5 seconds. Although it would be against the spirit of FIRST and stopping the flow of the game, it is legal at this point in the season before any rule updates to block the feeder station. This is something that if seen at a competition, would most likely be changed for the rest of the season.

[MrForbes]

The main reason I see for bothering with a floor pickup system, is for the advantage in autonomous. Last year it was a game decider for some teams that could get extra game pieces to score in autonomous.

[Speeder]

The main reason I see for bothering with a floor pickup system, is for the advantage in autonomous. Last year it was a game decider for some teams that could get extra game pieces to score in autonomous.

I'm not sure that Autonomous would be the only advantage. Even if all the 'bots were able to score at an 80% average, that could leave ~25 game pieces on the floor if they stayed on the Field.

[MrForbes]

Autonomous would not be the only advantage, but to me it appears to be the main advantage. Thinking of how many game pieces can be scored during a match, it's going to be a challenge for an alliance to use up all 45 in two minutes

[ThirteenOfTwo]

In answer to the original question...

I'd be very tempted to steal an intake from VRC Gateway. Those had to pick up cylinders (though those cylinders were much taller) and the intake design that ultimately emerged was efficient and lightning-fast. Rollers are king.

[pfreivald]

Autonomous would not be the only advantage, but to me it appears to be the main advantage. Thinking of how many game pieces can be scored during a match, it's going to be a challenge for an alliance to use up all 45 in two minutes

I'd rather spend ten seconds picking up disks I and my partners just missed with and re-shoot them, than spend twenty seconds driving down to the feeder, getting fed, and coming back to score.

[MrForbes]

If you can figure out how to pick up 4 discs in ten seconds, then you definitely should be doing that.

[DonRotolo]

Getting back to the thread title, how does one pick up a disk from the floor?

If we postulate a bot 24" wide, we can put rails near the floor (high enough to clear field imperfections) both fromt and back that funnel the disks to a given spot near robot center. Then, when a disk is detected in the right spot, rollers on both sides pop the disk up a bit. Something captures the disk and sends it off for further processing, maybe a pneumatic cylinder kicks it forward or something.

The rollers would be somewhat small diameter so they can catch the edge of the disk, maybe with little fingers of tubing instead of a solid foam/soft surface.

[Tom Line]

I'm not sure that Autonomous would be the only advantage. Even if all the 'bots were able to score at an 80% average, that could leave ~25 game pieces on the floor if they stayed on the Field.

I suspect you will see much much higher accuracy in this game. The game pieces aren't highly variable, and from 15 feet away there is one magic rule for these frisbees: the faster you fire it and the harder you spin it, the more accurate it will be, and the less drop (or more rise) you'll get.

Go back and look at 2006, where you were shooting into a vertically oriented goal instead of trying to bank things in off the backboard. That should be a more accurate representation of accuracy, and top robots that year were absolutely deadly (85-90%).

Prototyping shooters is absolutely key (if you plan on being a shooting bot).

[Bill_B]

How about if the rollers were inside out duct tape, like a lint roller? Just put on a new "belt" for each match, maybe?

[gabrielau23]

Personally, I think that picking up frisbees will be a waste of time unless you can do it in under 15 seconds. Given that at the beginning of the match there are VERY few frisbees to be collected, the first minute or so should be a period when you aren't even using your collection mechanism. The last 50 or so seconds, the room starts to populate with frisbees. However, you still have the endgame to play. Plus since each alliance can use 17 frisbees, I doubt teams will be able to make 4 runs back and forth. @ 20 seconds a run, thats 80 seconds, or 1 min and 20 seconds elapsed in tele=op. you still have 40 seconds left. This isn't even taking into account problems like defense and unexpected stuff that just "happens". I think 20 seconds will be a very fast turnaround time near the beginning of the game. 30 seconds left people should definitely start going for the climb.

[pfreivald]

I think 20 seconds will be a very fast turnaround time near the beginning of the game.

Yup. The times I chose were somewhat arbitrary in illustrating my point:

To be a great robot you can't just do things well--you also have to do them fast.

[jimwick]

There is a video this morning showing a shooter that seems to be able to make shots consistently from 50 feet.
So... one strategy might be to park at your feeder and just try to pump in shot after shot.
But.... if you miss, having a bot that could pick up from the floor and convert each miss to a score might be a big advantage.

And another strategy might be to shoot from the feeder into your end zone, have your teamates pick up and score. Lot quicker than driving around.

[AdamHeard]

I wasn't saying their shooter isn't precise, it could be for all I know. Just a word of caution to people trying to infer off the video. They only show a few shots, and with camera cuts. This could be a perfect shooter, it could be a mediocre one.

Teams really need to prototype and prove this for themselves to be truly confidant in their ability to perform.

I suspect you will see much much higher accuracy in this game. The game pieces aren't highly variable, and from 15 feet away there is one magic rule for these frisbees: the faster you fire it and the harder you spin it, the more accurate it will be, and the less drop (or more rise) you'll get.

Go back and look at 2006, where you were shooting into a vertically oriented goal instead of trying to bank things in off the backboard. That should be a more accurate representation of accuracy, and top robots that year were absolutely deadly (85-90%).

Prototyping shooters is absolutely key (if you plan on being a shooting bot).

Not only did they have higher accuracy in 06, but they were able to shoot MUCH faster than 2012 due to this.

[JohnFogarty]

One of my co-workers sent me this. The end of the video is more pertinent than the beginning.
http://www.youtube.com/watch?v=mDlHGhKHIdM&feature=youtu.be

[DonRotolo]

How about if the rollers were inside out duct tape, like a lint roller? Just put on a new "belt" for each match, maybe?I think you may find the glue loses its effectiveness after a few pickups, but definitely TRY IT. Dont take MY word for it!
Personally, I think that picking up frisbees will be a waste of time unless you can do it in under 15 seconds. Given that at the beginning of the match there are VERY few frisbees to be collected, the first minute or so should be a period when you aren't even using your collection mechanism.You mean, aside from the 10 or more already placed on the ground, and the autonomous shots that missed? I think you are wrong, but please dont take it personally.
Teams really need to prototype and prove this for themselves to be truly confidant in their ability to perform. Quoted for truth.

[DonRotolo]

One of my co-workers sent me this.
cool. I had seen that before, years ago, thanks for the reminder.

Just get something thin and flexible, like bumper cloth, hold it at the top and push out a thin sheet of something that rolls it out. Get under the disk, flip it up like a pancake, and repeat.

[nich]

One of the ideas that our team has tossed around is using some kind of vacume device to create suction in order to pick up the frisbee. We found that it took very little suction to do so. If you are using suction you can pick up the frisbee no matter what side is up. But we havn't discussed how to get the frisbee right side up in order to throw it yet.

[Lil' Lavery]

2. The rules state that you may block people at any point as long as they are not touching their pyramid, their safe zone next to the feeder station, and you do not pin them for more than 5 seconds. Although it would be against the spirit of FIRST and stopping the flow of the game, it is legal at this point in the season before any rule updates to block the feeder station. This is something that if seen at a competition, would most likely be changed for the rest of the season.

History would suggest the opposite. While FIRST has implemented many rules in order to encourage offense and protect offensive robots in the past few years, they haven't fully eliminated defense. Pushing and attempting to interfere with robot's transit across the field has been a long standing part of FRC games, and was a critical element in many games in which your alliance's human loaders are located on the far side of the field from where you score (2006 and 2011 being two recent examples).

[Timz3082]

I think people should check out this thera band material, it seems interesting... It might be perfect for frisbees!
http://www.chiefdelphi.com/forums/showthread.php?threadid=104281

[DonRotolo]

One of the ideas that our team has tossed around is using some kind of vacume device to create suction in order to pick up the frisbee. We found that it took very little suction to do so. If you are using suction you can pick up the frisbee no matter what side is up. But we havn't discussed how to get the frisbee right side up in order to throw it yet.
Two things to consider:
First, does it matter if they are upside down? Just saying you should test this first.
Second: Will vacuum still work of the disk is dusty/dirty/scuffed/smooshed a little?

[Toa Circuit]

I was thinking suction power for a bit... :D however, it takes our 5 horsepower shop vac being 3/8 of an inch away from the frisbee for it to lift it up. (Though once it is lifted, its staying put.) :(
So, considering that a robot definitely isn't going to generate a 5 horsepower vacuum, I think that its out of the question.

[depth_Finder]

vacuum cleaners have a rolling brush-like thing that helps suck things up. Although I think that a suction based system would be a bad idea, a rotating brush like thing might help the Frisbees onto a dustbin and keep them there. A vertical conveyor system that was common last year might be modified to lift Frisbees into the shooting mechanism.

[Ken Streeter]

I was thinking suction power for a bit... :D however, it takes our 5 horsepower shop vac being 3/8 of an inch away from the frisbee for it to lift it up. (Though once it is lifted, its staying put.) :(
So, considering that a robot definitely isn't going to generate a 5 horsepower vacuum, I think that its out of the question.

You might be surprised how much vacuum power an FRC robot can generate! Our 2010 robot used a vacuum to grab soccer balls. The same vacuum held a 2008 trackball quite solidly! A frisbee is a fair bit lighter than a trackball...

[Joon Park]

There is a video this morning showing a shooter that seems to be able to make shots consistently from 50 feet.

Can we perhaps see the video?

Also, I'm so glad robot in 3 days proved me wrong. It definitely taught me (and our team) to not confuse what a human can do and what a robot can do.

Getting back to the original topic though, it just goes to show how good of a game this is (props to the GDC). Time to prototype away!

[Littleboy]

How many balls in 2012 could 1717 pick up on average?
I imagine that this number will be about the same as the best robots that can pick up off the floor. Most teams won't be able to do close to that.
For the average team, human player loading is probably better than picking only a few off the ground.

[gabrielau23]

Two things to consider:
First, does it matter if they are upside down? Just saying you should test this first.
Second: Will vacuum still work of the disk is dusty/dirty/scuffed/smooshed a little?

According to video shot by Robot in 3 days, Shooting upside down has a negligible effect on distance. Accuracy also did not seem to be changed that much.

[ewrado]

My first idea was to find a way to make some kind of directional dry adhesive like the Stanford gecko robot http://nanolab.me.cmu.edu/projects/geckohair/ but although i thing it is possible would take a bit of time. So I started to think of something different that would do the same thing but simpler and sense there has been much talk of plungers/suction cups on chiefdelphi. I thought that a belt lined with many small plungers like the ones found in kids plastic gun toys may be a test worthy approach that may be able to lift a Frisbee from the top. Here is a link to the kind of suction cup i mean http://www.acrylicdesign.co.uk/products.php?action=showItems&params%5Bid%5D=260

[Mongai]

Our team gathered some insight on what would happen during a match. The human player has 40+ disks available to him or her. All that's available for most of the match to the player in terms of dispensing options are slots angled 45 degrees downward. Why should the human player just drop them through the slots onto the ground when they can be dropped into the robot directly? Sure, the frisbees can be thrown during the last 30 seconds of the match, but a person of average human height can only launch it over the wall at a risky angle. When I mean risky, I'm talking about the chance of losing accuracy because the disk will curve extremely (try throwing a frisbee 45 degrees upward into the air).

So the question really is: is picking up disks a) feasibly needed and b) worth the extra resources spent (match time, design time)? We thought otherwise.

[ferrari77]

There is a video this morning showing a shooter that seems to be able to make shots consistently from 50 feet.


I wasn't saying their shooter isn't precise, it could be for all I know. Just a word of caution to people trying to infer off the video. They only show a few shots, and with camera cuts. This could be a perfect shooter, it could be a mediocre one.

Response to those posts
I'm not sure which video the original comment was referring to but I know my team, 2583, posted a video yesterday showing 3 shots in a row within a foot and there is no editing. The first shot was a few feet off because we did not have our motor up to speed

The whole assembly is held together by hand so once it is mounted there will be less variation. I think this proves it is possible to have a very precise shooter

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

[3132MentorMike]

Two things to consider:
First, does it matter if they are upside down? Just saying you should test this first.
Second: Will vacuum still work of the disk is dusty/dirty/scuffed/smooshed a little?

The other thing to consider along this line is that the condition of disks will "evolve" over the course of the match. What may work for pristene disks at the start may not work as effectively as scuffs, dirt, gouges take their toll on disks....and vice versa

[pfreivald]

Response to those posts
I'm not sure which video the original comment was referring to but I know my team, 2583, posted a video yesterday showing 3 shots in a row within a foot and there is no editing. The first shot was a few feet off because we did not have our motor up to speed

The whole assembly is held together by hand so once it is mounted there will be less variation. I think this proves it is possible to have a very precise shooter

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

Well, that just answered a question we were going to answer tomorrow...

[Rishabhgadi]

Our team has been debating whether to use a shooter, or an arm. we have a prototype shooter but not an arm...Any input?

[MentorPrice]

I am just going to use my hands to pick up my frisbee I do not know about you guys. On a serious note I think that is a very challenging aspect of this years game.

[NanoCollins]

That's the easy part. The hard part is determining which way they are facing, and then how to flip them.

I think that picking up the frisbees is going to be a waste of time. There's a pretty small number sitting on the field, and the main way to get most of the frisbee's is through the feeding station.

However, I was thinking of a conveyer system which would have a little hatch open. Above the hatch would be a suction cup on a piston. It would go down, pick up the frisbee, then deposit it on the conveyer belt. Simple and fast. Doesn't solve the upside down issue though...

From what I have seen from other people's test, it does not matter all that much how they are flipped. Upside down or not (depending on your shooting system and the distance) they should fly fairly well.

[TheOtherGuy]

We bought a few suction cups at ACE for about $0.50 ea. along with a 3" ABS tube and some screws. It works quite well, actually, we just have to figure out how to get them from the suction cup to the hopper/shooter.

13422

[3044smike]

my team had a few ideas for flipping the disks but the one i like the most is a system where the disks actually come into the robot on the side :yikes: and then at the end there would be a little motor with two sticks with gear interactable teeth (don't know what they're called:confused: ) which would push the disk one way or the other based on a distance sensor. Any thoughts?!?

[3044smike]

From what I have seen from other people's test, it does not matter all that much how they are flipped. Upside down or not (depending on your shooting system and the distance) they should fly fairly well.

my team did a little testing and the upside down frisbees are fine and dandy for about 20 ft if you can throw it really hard with alot of spin but then it just drops like a rock::ouch::

flipping disks might still be important

[Libraryfanatic]

Sure it's much easier to go across the field to your alliance's feeding station and grab four discs, but think about it. After autonomous mode, you'd have 3 robots fighting for discs at the alliance feeding station. Sure there's enough slots for all three to be filled at the same time, but realistically, all three would not be fed at the same time, as the corner is very small and can barely fit all three robots there assuming that they are average sized. Plus, you have to worry about d-bots getting in your way and defending your feeding station. If you can pick up discs from the ground quick enough, it would be a HUGE advantage to you, and it would be beneficial during autonomous too, because then you can pick up discs from the ground and shoot them during auton for additional points (after you've shot your preloaded discs). And as for there being a lack of discs on the ground, the goals are fairly large but with the amount of defense that's possible, I can see a LOT of missed shots when a d-bot knocks a scoring bot off of alignment. We saw this last year with fender-shooters and d-bots.

There are three feeder stations-no competition at a given station.

[gabrielau23]

Well, I've flipped sides on collecting off the floor. I was pretty adamant in an earlier post about off-the-floor collection, but I have to say I agree with Joon now. Collecting will be a definite advantage, and in the final 30 seconds I could see teams scoring a lot from the discs thrown across the floor. Also, teams will probably miss at a higher rate then 85% unless they're "elite". And there's a reason there aren't that many "elite" teams. Picking up, I think, is going to be a very good thing to have. And yes, I everybody permission to make fun of me for flip flopping like this on this issue.

[3132MentorMike]

All good engineers proudly say, "I was wrong" if the data supports it.

[bobbyball]

But than how would you put them into your shooter?

just use a pully and belt system upwards

[GaryVoshol]

Our team has been debating whether to use a shooter, or an arm. we have a prototype shooter but not an arm...Any input?

So you're going only for the LOW GOAL? Maximum extension of an arm is ~5" short of the MIDDLE and ~20" short of the HIGH GOAL.

[bobbyball]

That's the easy part. The hard part is determining which way they are facing, and then how to flip them.

I think that picking up the frisbees is going to be a waste of time. There's a pretty small number sitting on the field, and the main way to get most of the frisbee's is through the feeding station.

However, I was thinking of a conveyer system which would have a little hatch open. Above the hatch would be a suction cup on a piston. It would go down, pick up the frisbee, then deposit it on the conveyer belt. Simple and fast. Doesn't solve the upside down issue though...

it will have to depend on the constictors of the belt shelfs

[Dan Oelke]

I created a prototype last night to sweep up the frisbee. This was using a 3/8" piece of plywood with about a 1/8" gap under the plywood. The frisbee would pretty consistently pop up onto the ramp.

There are 2 small wheels by the front edge. The idea being that those wheels will keep it from catching on the bump in the carpet. This also had 2 castors on the back edge of the ramp putting it about 4" off the ground.

http://www.youtube.com/watch?v=M7JbdU4h6F8

[Wyrdash]

I created a prototype last night to sweep up the frisbee. This was using a 3/8" piece of plywood with about a 1/8" gap under the plywood. The frisbee would pretty consistently pop up onto the ramp.

There are 2 small wheels by the front edge. The idea being that those wheels will keep it from catching on the bump in the carpet. This also had 2 castors on the back edge of the ramp putting it about 4" off the ground.

http://www.youtube.com/watch?v=M7JbdU4h6F8

It looks like this prototype is using hard brushes, something more sticky, or at least not dry nylon brushes will probably work better.

[Djur]

my team had a few ideas for flipping the disks but the one i like the most is a system where the disks actually come into the robot on the side :yikes: and then at the end there would be a little motor with two sticks with gear interactable teeth (don't know what they're called:confused: ) which would push the disk one way or the other based on a distance sensor. Any thoughts?!?

Those would be rack gears.

Why would you need to push the discs one way or another? Would you have two shooting mechanisms?

[Bruceb]

All this talk of weather to pick up from the floor or the feeder slots. I dont realy think it is that hard to pick up from the floor and I have devised a realy simple disk flipper that is integral to the conveyor and needs NO SENSOR. So, why not do both. If you are picking up from the floor then you have some kind of hopper. If you have a hopper then it should not be that difficult to collect from the feeder tubes. For that matter, if you can pick up from the floor you can pick up from the floor right in front of the feeder station. Put a sign on the bot for the human player " feed to floor".
I'll post a vid later showing the flipper.
Bruce

[Bruceb]

Additional note. Look at the flipper this way. If you lay the disk on the floor the leading edge of the disk is HIGHER if the disk is upside down. Yes, there is your hint.
Bruce

[vikesrock777]

Why would you need to push the discs one way or another? Would you have two shooting mechanisms?
I believe they were talking about pushing frisbees one way or another based on orientation. One side would be for frisbees that are already in the correct orientation while the other side would be for those that need flipping.

[Siri]

I believe they were talking about pushing frisbees one way or another based on orientation. One side would be for frisbees that are already in the correct orientation while the other side would be for those that need flipping.Should be simpler than that. If the disk is already on its side as they said, flipping them over is as conceptually simple as tilting one wall either direction. We got a prototype doing this over the weekend.

If you're trying to get them right-side up on the right hand side:

|.\..|
|..|.|
|./..| ...<< With the open side left, you open the left wall enough to tip it, then open right so it slides down and out.


|../.|
|.|..|
|..\.| ...<< With the open side right, you just open the right wall so it flops out. (May or may not need to kick the left wall inward, depending on bottom slope and exact geometry).


As far as floor pickup to vertical robot entry, I'll just leave this here: 2012 (http://baxterbombsquad.com/bbs/index.php/component/k2/itemlist/category/77-home).

[vikesrock777]

Should be simpler than that. If the disk is already on its side as they said, flipping them over is as conceptually simple as tilting one wall either direction. We got a prototype doing this over the weekend....
Ahh, I may have misunderstood when they said
where the disks actually come into the robot on the side.
I interpreted it as on the side of the robot as opposed to a frisbee in the orientation you drew out, though looking back that doesn't seem to make any sense :rolleyes: . Thanks for pointing that out.

[smclean1969]

Picking up the disks from the field will be a big advantage since there will be a lot of missed shots. That will be especially true in the playoffs where one or two teams are likely to be running back to the feeder zone and shooting. The third robot either plays defense or picks up the disks for some close up shots. I'm very confident the elite teams will be picking up the disks with ease.

[Herbblood]

My team is trying to work on the dispute as to whether or not to pick up off the floor. Our only problem to pick up is that our mentors prefer us to pick up from within the bumper frame, since we only have a small amt of space for it, we're no too sure on how to go with it. Any suggestions?

[F22Rapture]

That would depend entirely on how you're implementing your shooter and to a lesser extent your hopper.

[TimTheGreat]

I was thinking of using a goalie lacrosse stick. Putting it on pneumatics might make it go fast enough for it to scoop up the frisbee.

[MisterG]

I'm very confident the elite teams will be picking up the disks with ease.

I think that this is a thinking trap that we all fall into.

The elite teams do everything well; that's why they are elite teams. The question for the rest of us is whether adding a certain feature to our robot will be a good investment or not.

If the pickup allows you to acquire disks faster than a trip to the feeder station and it doesn't take too much away from other robot functions (size wise, resource wise, etc.) then it could be a plus.

In short, a good floor pickup that doesn't detract from other functionality would be a great thing. I personally don't see it as a must have in this years game.

Al G.

[CalTran]

Is floor pick up necessary? Back in Logomotion, floor pick up was thought to be a necessity. Team 1503 clearly showed everyone wrong. Their robot was hand fed until it was beefy enough to take on Einstein.

[SciBorg Dave]

Picking up a frisbee during seeding rounds would be a good idea. Maybe a lot of poor shooters.
During the elimination round there will not be that many disc on the floor to pick up because most of the shooters will be very good.

[falconmaster]

Here is a prototype we made that picks up frisbees and then the conveyor belt will act as a hopper where a pneumatic loader will pop the frisbee up and then a second piston will shoot the frisbee into the linear shooter that will be mounted on top of the conveyor belt
https://www.youtube.com/watch?feature=player_embedded&v=EUEg4TCwSd8
We can also use this arm to manual load from the human player

[gabrielau23]

Is floor pick up necessary? Back in Logomotion, floor pick up was thought to be a necessity. Team 1503 clearly showed everyone wrong. Their robot was hand fed until it was beefy enough to take on Einstein.

Well, tell that to 2537.....
We didn't think floor pickup would be neccesary either. Our autonomous was pretty darn good, and our minibot was no slouch, either. At Chesapeake and DC we got wrecked. Sometimes our own alliance members were wondering why we were just standing in front of the feeder station.

[cloudroth6]

the simplest solution i could think of was to pick them up like you'd pick up a basketball. just use rollers to suck it up.
that is what we are doing. taking what we learn from last years collector and putting it in to this years challenge.

[pfreivald]

that is what we are doing. taking what we learn from last years collector and putting it in to this years challenge.

And it even kind of not quite sort of works maybe the same-ish!

[Dinoyan]

How to pick up frisbee
http://www.youtube.com/watch?feature=player_embedded&v=YbTj3swVT18