Who is going over the bump? (A Pillar really)

[wilhitern1]

Originally Posted by ThirteenOfTwo

Q7) What is A5 * A6?
A7) Our team thinks it's between zero and ten seconds.

You left out in your analysis, how often is the bridge blocked and for how long. Think about all teams following your logic. How much bridge crossing is happening? It is just too dangerous for two bots to attempt to cross one bridge at once. Could it be done? yes, but not safely and quickly...

[BrendanB]

On day one our team did a little simulation of the game and from that we found that most of the moves our team members made were going across the field to get a ball from the human player in a competitive match (several scores from each alliance). With that being said we played around with whether or not going over the bump was necessary. We thought through it some more and came up with that if we just went over the bridge it would take a few seconds longer to cross but we are limited to 2 bridges: one of which is our own alliance's shared with 2 robots and a common bridge which is shared with 5 other robots 3 of which would want to stop us. Additionally, what if one of our robots got stuck on either bridge or takes much longer than us to cross then we are now stuck in rush hour traffic at the tolls, and in a game with so few balls what if for some reason we there are several ball in our opponents zone constantly?

Conclusion, crossing the bump while another design challenge will pay off in the long run in that we can move across the field over multiple access points to maximize our ability to play the game. To accomplish this we have angles on the front of our frame that allow us to glide over the bump with a 6wd.

We were also torn between a long and wide drivebase. A wide robot would give us a large gap for pickup and have an easier time fitting on the bridge but comes a the cost of having a high chance of tipping if not designed properly due to the shorter wheel base. A long robot takes away our ability for a large pickup but is safer at going over the bump and fitting in our ball pickup and shooter.

Conclusion, make a long robot with a drop down harvester at full width and is heavily weighted towards one side to fit on the bridge for a triple balance.

Pretty lengthy but this is what our team came up with in the end. The biggest thing is making sure your team can accomplish what you intend to build.

Good luck!

[MrForbes]

We don't know if we're going over the barrier. Hopefully we'll know in a week or two. But we are going to make the robot so it can attempt it.

[tsaksa]

The students really wanted to go over the bump, so that is what we are doing even though some of the mentors still think this puts the schedule at risk. Our team has traditionally done track drive which takes more machining, expense, and tweaking, than other drive bases. Last year they had more difficulty (than usual) getting everything done on time. So for practice over the summer they decided to build a demo-bot using regular wheeled kit-bot parts. From that experience they seemed ready to build a normal wheeled robot this year so they could focus more on the other parts of the challange. Than they saw the barrier that seemed to call out for tracks. What to do!

What we came up with is a wheeled robot that still has some track like characteristics. In stead of chains in some places it has timing belts. These belts are supported in such a way as to act enough like a track to hopefully allow the robot to avoid high centering when crossing the barrier. It is not a real track and will not be good enough to use much, but hopefully it will work just enough to navigate the barrier.

If everything works out, and that still seems like a big if right now, we will have a robot that drives around most of the time like a wheeled robot with just a little assistance from the pseudo tracks only when needed. We are using custom wheel mounting plates, but believe the design could ultimately be adapted to a kit of parts frame to allow teams to experiment with track drive without a lot of machining or expense.

The design is lower cost than traditional track drive and requires less machining because it uses the standard Gates timing belts with off the shelf pulleys. But that means you only get part of the advantages of track drive. The students are calling this variation Corvallis Spartan Drive because they do not know if it has ever been tried before. We will need to post some pictures if we ever get it working. Too soon to say if it has any advantage, or will even be competitive, but that is what happens when you try something new.

[Larry Lewis]

I know for our team being able to get anywhere on the field rather quickly has always been beneficial to us. This is a lesson we have learned by looking at the robots that you routinely see in the finals at championships. Generally speaking they are able to easily transition from different zones of the field quickly. This allows them to acquire game pieces quickly and score them regardless of where the game pieces are on the field.

Therefore, I think that we will attempt that approach again this year but I can see where robots that can't transition the barrier are useful. If they are a really good shooting robot they can stay in their scoring zone while other robots on their alliance cross over to the other side to feed them ball.

[Donut]

Our robot is designed to focus on bridge crossing since we've been able to cross it fairly quickly and easily with a prototype mechanism we built. The plan is to build in the ability to cross the barrier as a last resort but it's not a part of our regular strategy and would probably cause damage to our drivetrain if used on a regular basis (we're using mecanum wheels that we don't want to ram into the barrier repeatedly).

Really we're hoping to end up in a feeder/shooter alliance setup and avoid crossing the barrier as much as possible for all teams involved.

[pfreivald]

Originally Posted by Donut

Our robot is designed to focus on bridge crossing since we've been able to cross it fairly quickly and easily with a prototype mechanism we built. The plan is to build in the ability to cross the barrier as a last resort but it's not a part of our regular strategy and would probably cause damage to our drivetrain if used on a regular basis (we're using mecanum wheels that we don't want to ram into the barrier repeatedly).

Really we're hoping to end up in a feeder/shooter alliance setup and avoid crossing the barrier as much as possible for all teams involved.

I think that all teams will avoid crossing the barrier as much as possible -- but really great teams will be able to cross it as necessary.

We're going octocanum this year, but have something in place to keep the shock of the barrier off of the drive train.

[ThirteenOfTwo]

Originally Posted by wilhitern1

You left out in your analysis, how often is the bridge blocked and for how long. Think about all teams following your logic. How much bridge crossing is happening? It is just too dangerous for two bots to attempt to cross one bridge at once. Could it be done? yes, but not safely and quickly...

Ooh, you're right.

I suppose we should conclude that our answer to that question was "never".

1) Not very much bridge crossing is happening.*
2) Opponents are not stupid enough to block your bridge and risk a technical foul.
3) Even if they block your bridge, the Coopertition bridge is open and blocking you there is a technical foul under the blockading rule.
4) That leaves two allies who can block your bridge at a time.
5) One of those allies will likely be playing defense and will thus cross midfield only once, right at the beginning.
6) That leaves one ally who can block your bridge at a time.
7) If both of you cross the bridge at the same time, neither of you is scoring.
8) If neither of you is scoring, you are losing.
9) You should never be crossing the bridge at the same time as your partner.

*This could be a flawed assumption, in which case the bump is much more important than we think. However, all of our simulations and logic back this up.

Originally Posted by BrendanB

On day one our team did a little simulation of the game and from that we found that most of the moves our team members made were going across the field to get a ball from the human player in a competitive match (several scores from each alliance). With that being said we played around with whether or not going over the bump was necessary. Additionally, what if one of our robots got stuck on either bridge or takes much longer than us to cross then we are now stuck in rush hour traffic at the tolls, and in a game with so few balls what if for some reason we there are several ball in our opponents zone constantly?

This sounds like sketchy logic. Did your simulating players try bouncing the ball from the inbounder station? What about dedicating one robot on the other side of the field to play defense AND to return scored balls?

Originally Posted by BrendanB

Additionally, what if one of our robots got stuck on either bridge or takes much longer than us to cross then we are now stuck in rush hour traffic at the tolls

In your simulations, did you ever come across a situation where more than one robot needed to use a bridge at a time? "We'd be screwed if our partner breaks down loses some validity because if your partner breaks down on the bridge (thus rendering it unbalanceable) it's probably already over.

Originally Posted by BrendanB

and in a game with so few balls what if for some reason we there are several ball in our opponents zone constantly?

The flaw here is "for some reason". Think about the reasons this would happen.

Q1) Why do they have a lot of balls?
A1) There are two possibilities.

P1) They are receiving a lot of balls.
P2) They are not scoring any balls.

C1) If P2 is true, they are losing.

Q2) From where are they receiving these balls?
A2) There are several possibilities.

P3) They are receiving them from their feeder station faster than they can score them.
P4) They are receiving them from the ground on your side of the field faster than they can score them.
P5) Some combination of P3 and P4

C2) If P3 is true, you are scoring at a faster rate than they are and therefore winning.
C3) If P4 is true, your aim is bad and their defensive robot is amazing at stealing rebounds.
C4) If P5 is true, the result is likely a combination of C2 and C3.

O1) If P4 is true, stop that defensive robot! (Or, alternately, build a robot that is accurate.)
O2) If P5 is true, stop that defensive robot! (Or, alternately, build a robot that is accurate.) (Or, if P3 > P4, don't worry!)

C4) If they have a lot of balls on their side of the field, either you are winning or one robot is responsible for their success or your alliance is so wildly inaccurate that you have no hope of winning anyways.

[team222badbrad]

We plan to slide right over the bump like it wasn't even there...

[pfreivald]

Originally Posted by team222badbrad

We plan to slide right over the bump like it wasn't even there...

Hoverbot? Or teleport bot?

Originally Posted by team222badbrad

We plan to slide right over the bump like it wasn't even there...

Are you gonna go Wildstang 2004 on us?

[BrendanB]

Originally Posted by ThirteenOfTwo

\\

This sounds like sketchy logic. Did your simulating players try bouncing the ball from the inbounder station? What about dedicating one robot on the other side of the field to play defense AND to return scored balls? In your simulations, did you ever come across a situation where more than one robot needed to use a bridge at a time? "We'd be screwed if our partner breaks down loses some validity because if your partner breaks down on the bridge (thus rendering it unbalanceable) it's probably already over. The flaw here is "for some reason". Think about the reasons this would happen.

Q1) Why do they have a lot of balls?
A1) There are two possibilities.

P1) They are receiving a lot of balls.
P2) They are not scoring any balls.

C1) If P2 is true, they are losing.

Q2) From where are they receiving these balls?
A2) There are several possibilities.

P3) They are receiving them from their feeder station faster than they can score them.
P4) They are receiving them from the ground on your side of the field faster than they can score them.
P5) Some combination of P3 and P4

C2) If P3 is true, you are scoring at a faster rate than they are and therefore winning.
C3) If P4 is true, your aim is bad and their defensive robot is amazing at stealing rebounds.
C4) If P5 is true, the result is likely a combination of C2 and C3.

O1) If P4 is true, stop that defensive robot! (Or, alternately, build a robot that is accurate.)
O2) If P5 is true, stop that defensive robot! (Or, alternately, build a robot that is accurate.) (Or, if P3 > P4, don't worry!)

C4) If they have a lot of balls on their side of the field, either you are winning or one robot is responsible for their success or your alliance is so wildly inaccurate that you have no hope of winning anyways.

Since we did not have a full field we did a table top simulation. I also got a chance to play with the field that morning. Our simulation assumed that no balls would be thrown or bounced over the field (we may play around with this but unless you have an inbound station, bridge, and bump you can't rule it possible).

While it may seem like, "sketchy logic" it is perfectly resonable to me. We participate in a week one regional where many teams are just getting up and running and driver skill is on the lower side unless they have practiced. All you need is one robot to have a hard time going over the bridge and you have yourself a mess. I never assume that my robot or the robots at my event will be stellar at one thing or doing something will be easy like crossing the bump or even driving. That isn't mean or saying everyone else is horrible, I'm just saying that I'm not going to assume my regional has 50 robots that can cross a bridge in 2 seconds. Following that same train of though while having one robot to funnel balls across the field is a good partner, I'm not going to assume that robot will be at my regional or in my matches. If I designed around this what am I going to do when I'm stuck with 2 other robots just like me or can't do anything?

If a partner breaks down on the ramp/is hanging off the side I'm not going to touch it unless it is absolutaly necessary out of respect for that teams robot. I will however design to go over the bump so I don't have to worry about the bridge until the end. Our design should take around 1 second to cross the bump, unless I can make a bridge cross that fast including dropping the bridge I'd do it. Again I'm not going to assume the bridge is open the entire match. To say my partners are going to stay in one zone is out there because I can't control what they do.

Assuming you are correct and they are outscoring me does that mean I stop scoring? The way I see this game is that there are 18 balls which equates to 3 per robot which is all I can have. Each inbounder can hold 2 balls which means up to 6 on each side. If you take a moment and think that over and over you'll realize that there aren't going to be a lot of balls on the field (I'm not saying 3 balls will be in each robot at all times and each inbounder will have 2 at all times). Collecting the few balls that are on the field will be a huge factor in this years game which is another reason why we see the bump as a strategic point. If it is faster to cross the bump then the bridge we will cross the bump as much as we can.

In short I'm not assuming the bridge is open all the time and I'm not going design my robot around what I think other teams can do.

[Heretic121]

Originally Posted by wilhitern1

You left out in your analysis, how often is the bridge blocked and for how long. Think about all teams following your logic. How much bridge crossing is happening? It is just too dangerous for two bots to attempt to cross one bridge at once. Could it be done? yes, but not safely and quickly...

This is where strategy comes to play... talking to your alliance partners before matches is key... you should never have a bridge jam!

[Chris Fultz]

One thought to consider with the bridges is that the center bridge 'exit' location is pretty close to the key. If you come off of the bridge too quicly you have a high risk of contacting an opposing alliance robot that is in that zone.

[ThirteenOfTwo]

Originally Posted by BrendanB

Since we did not have a full field we did a table top simulation. I also got a chance to play with the field that morning. Our simulation assumed that no balls would be thrown or bounced over the field (we may play around with this but unless you have an inbound station, bridge, and bump you can't rule it possible).

That's a hugely important incorrect assumption. When running your simulations, you should always simulate the real game, not change the rules to your liking. There's no point in simulating a game that you won't be playing.

Originally Posted by BrendanB

All you need is one robot to have a hard time going over the bridge and you have yourself a mess.

Well, not quite. All you need to have a mess is one robot that breaks down on your bridge (as discussed earlier, game over already). If it has a hard time, you can and should design a strategy to minimize its crossings.

Originally Posted by BrendanB

I never assume that my robot or the robots at my event will be stellar at one thing or doing something will be easy like crossing the bump or even driving. That isn't mean or saying everyone else is horrible, I'm just saying that I'm not going to assume my regional has 50 robots that can cross a bridge in 2 seconds.

Neither did we; I mentioned earlier that our analysis allots quite a chunk of extra time for bridge crossing compared to bump crossing. Just as you can't assume bridge crossing time will be zero, I can't assume bump crossing time will be. I think a five second differential for every crossing is probably reasonable. You can go higher... but as per our team's analysis, a ten-second differential (whoa!) is still only ten wasted seconds per match at most.

Originally Posted by BrendanB

Following that same train of though while having one robot to funnel balls across the field is a good partner, I'm not going to assume that robot will be at my regional or in my matches.

I should hope that there will be a robot like that at your competition--all it has to be able to do is pick up balls and throw them across the midfield gap for you to pick them up.

Originally Posted by BrendanB

If I designed around this what am I going to do when I'm stuck with 2 other robots just like me or can't do anything?

Start making bounce passes from the inbounder station or only send one at a time to cross the bridge.

Originally Posted by BrendanB

If a partner breaks down on the ramp/is hanging off the side I'm not going to touch it unless it is absolutaly necessary out of respect for that teams robot. I will however design to go over the bump so I don't have to worry about the bridge until the end.

See, the difference here is that you think you have to cross the bump in order to not have to worry about the bridge, and I think you can ignore them both altogether if your scoring mechanism is accurate enough. What it boils down to is that I assumed people would be throwing balls across the field and you assumed that they would not.

Originally Posted by BrendanB

Our design should take around 1 second to cross the bump, unless I can make a bridge cross that fast including dropping the bridge I'd do it.

See again my earlier point: A bump crossing mechanism makes any robot better, and of course you'd rather cross the bump than the bridge if you can, but for some teams it may not be worth the requisite engineering time.

Originally Posted by BrendanB

Again I'm not going to assume the bridge is open the entire match. To say my partners are going to stay in one zone is out there because I can't control what they do.

You can't? This is where pre-match discussions need to come in. Why wouldn't you tell your alliance partners to default to the most efficient possible configuration?

Originally Posted by BrendanB

Assuming you are correct and they are outscoring me does that mean I stop scoring?

In none of my scenarios were they outscoring you, expect the one in which you are inaccurate. If you are inaccurate you haven't even started scoring, so I'm not sure this makes sense...

Originally Posted by BrendanB

The way I see this game is that there are 18 balls which equates to 3 per robot which is all I can have. Each inbounder can hold 2 balls which means up to 6 on each side. If you take a moment and think that over and over you'll realize that there aren't going to be a lot of balls on the field (I'm not saying 3 balls will be in each robot at all times and each inbounder will have 2 at all times). Collecting the few balls that are on the field will be a huge factor in this years game which is another reason why we see the bump as a strategic point. If it is faster to cross the bump then the bridge we will cross the bump as much as we can.

No, you won't... you will cross the bump as much as you need to. Crossing midfield for its own sake is silly because it's wasted time. If you think you need to retrieve balls from the other side of the field, why not have one robot stay on the other side of the field to retrieve them?

Originally Posted by BrendanB

In short I'm not assuming the bridge is open all the time and I'm not going design my robot around what I think other teams can do.

We didn't assume it either; we simulated it. Then, based on our simulation, we decided what was and what wasn't an important design objective and decided to postpone pursuing a seemingly unimportant objective that would have required meaningful investment of time on our part.