View Full Version : How can we drive on this new floor?
We cant use OMNI's or traction wheels, only the KOP wheels. any ideas of how to drive?
Jay H 237
03-01-2009, 13:10
The same way you would as if you were driving on snow or ice. It is possible and start off slow to get going.
I think the teams in the snow belt areas will have an advantage with the 2009game :D
Seat Ninja
03-01-2009, 17:36
My guess would be to accelerate slowly to prevent burnouts, and distribute your weight on your back wheels.
Well, it's obvious that every wheel touching the ground will need to be powered (otherwise you're foolishly wasting precious traction) but beyond that I don't have many ideas. Would any non-tank drive style drive systems help? Also, am I wrong in assuming that traction force is just a function of coefficient of friction (cannot be changed) and mass (going to be 120 for everybody)? Does increasing the surface area have any affect?
There are even a few more factors than that in optimizing traction.
However, to put it simply:
Friction is a function of normal force (weight).
Traction is a function of normal force AND surface area.
Would there be any advantage in using crab drive or another drive system like that? Or would a car like drive system be better for towing the trailer around?
drkiraco
03-01-2009, 18:31
<R06> says it all. You may use only the KOP wheels.
We cant use OMNI's or traction wheels, only the KOP wheels. any ideas of how to drive?
gorrilla
03-01-2009, 18:33
<R06> says it all. You may use only the KOP wheels.
they dident say that we couldent angle the wheels............
<R06> says it all. You may use only the KOP wheels.
A holonomic drivebase would still be possible even with just the KOP wheels. The coefficient of friction is low enough that a swerve base or kiwi drive would both be possible, since the wheels slide easily enough.
gorrilla
03-01-2009, 18:34
A holonomic drivebase would still be possible even with just the KOP wheels. The coefficient of friction is low enough that a swerve base or kiwi drive would both be possible, since the wheels slide easily enough.
thats exactly what i was thinking
Swerting
03-01-2009, 18:35
I think it may be good to have weight on the driving wheels so that there is plenty of traction for turning and changing direction.
There is another rule stating that the wheel's rotation MUST BE inline with movement of the bot, so that the axle the wheel is on is perpendicular to the bot and wheels movement
markulrich
03-01-2009, 18:43
For tank drive systems, what do you guys think about 4 vs 6 coplanar, evenly spaced wheels? Would there really be any difference?
Has anyone felt the new playing surface? How much does it cost? Would plane old hallway concrete be a suitable substitute?
gorrilla
03-01-2009, 18:43
There is another rule stating that the wheel's rotation MUST BE inline with movement of the bot, so that the axle the wheel is on is perpendicular to the bot and wheels movement
could you find this and link to it?
Bryan Herbst
03-01-2009, 18:45
The same way you would as if you were driving on snow or ice. It is possible and start off slow to get going.
I think the teams in the snow belt areas will have an advantage with the 2009game :D
I have to agree.
To teams that have no experience driving in snowy/icy conditions- trying to start out going fast (even middle-range speeds) can result in either your tires just sitting there spinning, followed by a jerk forward, or the car/bot sliding out of control.
The fewer wheels you have will actually increase you traction because you will have more weight on the drive wheels we are thinking about having a front wheel drive with one more wheel in the back. We would use a tank drive to be able to turn with the trailer.
could you find this and link to it?
<R06>
ROBOTs must use ROVER WHEELS (as supplied in the 2009 Kit Of Parts and/or their equivalent as provided by the supplying vendor) to provide traction between the ROBOT and the ARENA. Any number of ROVER WHEELS may be used. The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub). No other forms of traction devices (wheels, tracks, legs, or other devices intended to provide traction) are permitted. The surface tread of the ROVER WHEELS may not be modified except through normal wear-and-tear. Specifically, the addition of cleats, studs, carved treads, alterations to the wheel profile, high-traction surface treatments, adhesive coatings, abrasive materials, and/or other attachments are prohibited. The intent of this rule is that the ROVER WHEELS be used in as close to their “out of the box” condition as possible, to provide the intended low-friction dynamic performance during the game
Seems to me it says nothing about requiring the wheels be oriented so MOTION is in the direction of the tread, only that they be mounted with tread touching the floor (i imagine to foil us lateral thinkers that might try to cantilever the wheels and use the flat face of them to try to move forward)
im not sure how many of you out there live in the lovely area that gets a lot of snow. but im in new carlisle indiana. and we just get dumped on constanlty. one thing i've notices is front wheel drive cars tend to do better. you can power out of the slide (if you are able to do a car-like steering). and i have noticed that rear wheel drive causes a lot of fishtailing.(maybe not with the trailer). but the best by far is four-wheel drive. it gives you power to all four wheels. team 2197 is without a doubt going with four wheel drive. good luck to everyone. i hope you all have fun with lunacy
writchie
03-01-2009, 18:54
There is another rule stating that the wheel's rotation MUST BE inline with movement of the bot, so that the axle the wheel is on is perpendicular to the bot and wheels movement
If you are talking about R06 I don't read it that way. The tread must be in contact with the ground and the wheel axis must be parallel to the ground and through the wheel hub. However, this does not imply "inline. At first read, I do not see what rule would prohibit a crab steering configuration unless I missed something.
dtengineering
03-01-2009, 19:00
One thing to watch for when designing the robot is turning ability when using skid steering (tank drive).
If you notice the coefficients of friction for the wheels, you will see that inline dynamic is .05 while transverse dynamic is .10. In a skidding situation the wheel will be twice as resistant to going sideways as it is to going forwards....
That means, as I read it, that your wheel is not going to want to slide sideways very easily if you use 4wd tank drive.
In fact anyone planning on going with skid steering will probably want to really look at using at least six wheels to minimize the amount of skidding required in a turn.
Jason
thefro526
03-01-2009, 19:02
I think the intention of R06 is to prevent teams from using the non-treaded portion of the wheel as a traction surface, not rule out Holonomic or Kiwi type drive systems.
thefro526
03-01-2009, 19:12
As I think more and more about this game I've come to a conclusion.
No matter what you do, you'll still be working with limited traction. Instead of worry about getting more traction, why not just embrace the lose of traction?
I think the question of "how can we drive on this new floor?" should be posed to the drivers. It'll require quite a bit of finesse and skill to master the lack of traction but those who do will be at a huge advantage.
Perhaps you should have your drivers research things like Drifting, ice driving, and driving in the rain to get some ideas. I personally, dabble in a bit of drifting myself so I'd like to see if I can apply that experience to driving a robot on this surface.
I think the solution for driving on this new floor will be more likely on Software : http://en.wikipedia.org/wiki/Traction_control_system.
Gotta think how does the algorithm work on a 4W drive vehicle works......
Cheers,
Marcos.
As I think more and more about this game I've come to a conclusion.
No matter what you do, you'll still be working with limited traction. Instead of worry about getting more traction, why not just embrace the lose of traction?
I think the question of "how can we drive on this new floor?" should be posed to the drivers. It'll require quite a bit of finesse and skill to master the lack of traction but those who do will be at a huge advantage.
Perhaps you should have your drivers research things like Drifting, ice driving, and driving in the rain to get some ideas. I personally, dabble in a bit of drifting myself so I'd like to see if I can apply that experience to driving a robot on this surface.
i totally agree with that. theres no way to increase traction. and i also do some drifting myself. main thing is you cant give a lot of power right away. otherwise you will just spin your tires(wheels). it will require some skills by the drivers
writchie
03-01-2009, 19:21
One thing to watch for when designing the robot is turning ability when using skid steering (tank drive).
If you notice the coefficients of friction for the wheels, you will see that inline dynamic is .05 while transverse dynamic is .10. In a skidding situation the wheel will be twice as resistant to going sideways as it is to going forwards....
That means, as I read it, that your wheel is not going to want to slide sideways very easily if you use 4wd tank drive.
In fact anyone planning on going with skid steering will probably want to really look at using at least six wheels to minimize the amount of skidding required in a turn.
Jason
It may be even worse than it first appears. Unless you've already broken traction, skid steering has to overcome what could be as high as the static transverse friction (if you want to turn as you are starting out). At first glance it appears 6 wheel drive raised center is the way to go for skid steering (whether tank mode or joystick mode on the controls). Also the abnormally high differences between static and dynamic coefficients is probably worth a bit more analysis. In past years static vs. dynamic and inline versus transverse were relatively minor differences that could often be more or less ignored. Not so this year. Don't count on being able to control the bot until you do the math ;)
Branden Ghena
03-01-2009, 19:23
Seems to me it says nothing about requiring the wheels be oriented so MOTION is in the direction of the tread, only that they be mounted with tread touching the floor (i imagine to foil us lateral thinkers that might try to cantilever the wheels and use the flat face of them to try to move forward)
Darn, the first thing I thought was that we should make a walker bot using the wheels sideways. :D
You can add more traction to the wheels by adding downward force onto the wheels. Any additional weight would be dispersed by the wheels regardless of where it's centered unless the robot has another point of contact with the ground (or if it goes all the way through the trailer). So you could add force with springs or with pneumatics, thereby adding traction.
But I'm wondering, could you propel yourself by say, pushing off of walls? There have got to be some ways to move outside of the traditional wheels on the ground that are still legal within the competition constraints.
robodude03
03-01-2009, 19:44
As I think more and more about this game I've come to a conclusion.
No matter what you do, you'll still be working with limited traction. Instead of worry about getting more traction, why not just embrace the lose of traction?
I think the question of "how can we drive on this new floor?" should be posed to the drivers. It'll require quite a bit of finesse and skill to master the lack of traction but those who do will be at a huge advantage.
Perhaps you should have your drivers research things like Drifting, ice driving, and driving in the rain to get some ideas. I personally, dabble in a bit of drifting myself so I'd like to see if I can apply that experience to driving a robot on this surface.
You are absolutely right. Drivers will need to train themselves and adapt to the slick situation. Additionally, I believe that programming will also have a big part in this years game (i.e. Traction Control). Our team will be doing a lot of testing in terms of the best method of acceleration for our robot, inputting that into our system and applying that to our controls.
Zyck_titan
03-01-2009, 19:47
It seems to me that heavier Bots will do a lot better, and I'm sure everyone will be maxed out on weight, so the question is how do you add weight....without adding weight....;) *Hint hint Nudge nudge*
thefro526
03-01-2009, 19:50
It seems to me that heavier Bots will do a lot better, and I'm sure everyone will be maxed out on weight, so the question is how do you add weight....without adding weight....;) *Hint hint Nudge nudge*
That's an easy assumption to make.
But what about momentum? Changing directions with a heavier bot it's going be be a bit harder I think... Or maybe not, I'd like to get some input on this.
Enigma's puzzle
03-01-2009, 19:50
Sorry to break it too you all but tank drive is going to be a pain, Tank drive has an axis of rotation around the center of the robot. BUT you have a trailer, you are going to be dragging the trailer wheels side ways it will be a battle.
Also the trailer anytime that you backup you will jack knife almost immediately, then you will be unable to control your backwards motion, turning with the trailer.
germanystudent
03-01-2009, 19:53
Haha! i bet everyone in the north U.S. has a bit of an advantage with this above average snowfall. similar to snow and ice, you would want to start slow, and don't lock up your wheels... try pulsing them like in an anti lock break system, or just slowing down. weight to increase traction is also double sided, sure you can accelerate faster, but then again you take further to stop weather that be sliding or slowing down. It pays to drive in Idaho!
mray1031
03-01-2009, 19:53
Anyone thinking about workarounds to the traction problem?
If the total traction is a function of your friction, surface area, and weight, and friction is fixed- shouldn't we be looking at surface area and weight?
Surface area is easy- more wheels (I don't think there's a limit)
But weight? could be interesting. Any ideas?
thefro526
03-01-2009, 19:56
Sorry to break it too you all but tank drive is going to be a pain, Tank drive has an axis of rotation around the center of the robot. BUT you have a trailer, you are going to be dragging the trailer wheels side ways it will be a battle.
Also the trailer anytime that you backup you will jack knife almost immediately, then you will be unable to control your backwards motion, turning with the trailer.
In theory though, you should be able to drag the trailer around because the wheels are the same non-grippy wheels that are on the robot. I'd also venture to guess that a trailer weighs substantially less than a robot, but then again I'm not entirely sure...
mray1031
03-01-2009, 19:56
oh man. people are posting fast i guess my last one was outdated...
since people are already thinking about pluses and minuses of weight anyone consider varying the "weight" of your robot?
Enigma's puzzle
03-01-2009, 20:00
To add wieght with without adding wieght you could always fill up your trailer
dmlawrence
03-01-2009, 20:01
In theory, one could construct giant omniwheels using each Rover Wheel as one of the horizontal rollers...
mray1031
03-01-2009, 20:02
Yeah by adding to the trailer you gain mass - just not on your drive wheels. You'll gain momentum, just not traction. Plus you're giving the other team points.
Enigma's puzzle
03-01-2009, 20:06
Traction is detirmined by pounds per square inch, so more wheels is taking away from your traction on each one, so every non-drive wheel is a waste of potential traction theoretically.
All wheels have the same coefficient of friction so because of the wieght differential the trailer will move, but slowly
Dracco1993
03-01-2009, 20:10
I am a driver from the FRC team 1741. I like the idea of drifting to gain more control of the robots. My only concern is that with the trailer it will be very hard to turn quickly. Can you get enough speed to drift in the arena?
:ahh:
Branden Ghena
03-01-2009, 20:14
In theory, one could construct giant omniwheels using each Rover Wheel as one of the horizontal rollers...
While a great idea...
Originally Posted by FRC 2009 Game Manual
<R06>
ROBOTs must use ROVER WHEELS (as supplied in the 2009 Kit Of Parts and/or their equivalent as provided by the supplying vendor) to provide traction between the ROBOT and the ARENA. Any number of ROVER WHEELS may be used. The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub). No other forms of traction devices (wheels, tracks, legs, or other devices intended to provide traction) are permitted. The surface tread of the ROVER WHEELS may not be modified except through normal wear-and-tear. Specifically, the addition of cleats, studs, carved treads, alterations to the wheel profile, high-traction surface treatments, adhesive coatings, abrasive materials, and/or other attachments are prohibited. The intent of this rule is that the ROVER WHEELS be used in as close to their “out of the box” condition as possible, to provide the intended low-friction dynamic performance during the game
I think the emphasized part of R06 might disallow that, along with the general logistics of building giant omniwheels. :D
darkpheonix
03-01-2009, 20:19
In response to adding weight without adding weight, you could add fans to the bottom of the robot to draw in air from the bottom, just like those cars that can climb walls. Don't know how much it would increase the normal force though.
There seems to be a lot of confusion going on about some physics, I think this should help:
ForceFriction = Mass*Gravity*CoefficientOfFriction
surface area is not a part of the calculation, as increased surface area means the normal force (force of the ground pushing up) is more spread out.
Force = Mass*Acceleration
thus
Mass*Gravity*CoefficientOfFriction = Mass*Acceleration
mass cancels
Gravity*CoefficientOfFriction = Acceleration
which shows that maximum acceleration is a function of nothing more than Gravity and the Coefficient of Friction, neither of which can be changed. Any team that powers all wheels in contact with the ground will have a top acceleration of .588 assuming no slippage (.06 cof) or .49 with slipping (.05 cof). The top speed assuming you drive straight across the regolith is approximately 2.2 meters per second, or about 5 miles per hour. The trip would take about 4.5 seconds.
By the way, where is everybody getting the coefficients of friction? How accurate are they? Doe anybody know the cof of the wheels on carpet?
Andrew Bates
03-01-2009, 20:22
What if your robot had two drive trains? One in the forward direction and one in perpendicular to the forward direction. You lifted one drive train to use the other. Kinda like the trucks that are made to drive on the road and railroad tracks.
Zyck_titan
03-01-2009, 20:23
That's an easy assumption to make.
But what about momentum? Changing directions with a heavier bot it's going be be a bit harder I think... Or maybe not, I'd like to get some input on this.
Well what if you could control how much you wanted, you know actively change from a heavier bot to a lighter bot in the middle of a match, you could start moving by being light, "make" yourself heavy if you don't want to be pushed etc.
I do have an idea for this, but I'm trying to stimulate the thinktank, see what else people come up with :)
gorrilla
03-01-2009, 20:34
car steering might make a comeback this year......seeing as how tank drive wouldent turnas well,(drifting around with the trailer would take some getting use to though)
dmlawrence
03-01-2009, 20:37
In theory, one could construct giant omniwheels using each Rover Wheel as one of the horizontal rollers...
<R06>
...The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub)...
Originally posted by Branden Ghena
I think...R06 might disallow that, along with the general logistics of building giant omniwheels.
I believe R06 would only apply to the wheels being "used" - that is, the ones touching the ground. The configuration would then be legal, since the wheels not satisfying R06 would not be in use.
Such an omniwheel would have nearly triple the traction of a single Rover Wheel.
Mike8519
03-01-2009, 20:38
In response to adding weight without adding weight, you could add fans to the bottom of the robot to draw in air from the bottom, just like those cars that can climb walls. Don't know how much it would increase the normal force though.
Reading R06... It says to me that no devices at all (making contact or not) may be used as a "traction device". The intent is that all robots have similiar traction on the surface. Increasing your force via suction would be a "traction device" in my mind
Some quick suggestions off the top of my head:
Giant Flywheel. For resistance to bumps and bruises, spin up a giant flywheel to help you conserve momentum.
Traction Control. This was mentioned before, but I'll summarize the idea of a traction control system. In a car, you compare the speed of the powered wheels to the unpowered wheels. If there is a difference, that means one set is slipping while the other isn't. The computer cuts a bit of power to the driven wheels until both spin at the same rate. For a 4-wheel vehicle, you can use an accelerometer to get the differential measurement.
Mariocart for Driver Training. Mariocart invented drifting. Blue sparks may be a safety hazard, though.
Pneumatics as Thrusters. Satellites use gas thrusters to control direction. If the surface is low enough friction, this could act as a nice stability augmentation.
Antilock Brakes. Antilock brakes don't necessary cut down on braking distance, but rather they give you control WHILE braking. If you slam on a car that doesn't have antilock brakes and you lock the wheels, the car basically keeps moving in the direction of its momentum vector, regardless of which direction the steering wheel is pointing (conservation of momentum). With antilock brakes, you're switching between static friction and dynamic friction, and in a nut shell, this gives you some control (ie allows you to swerve around the object you're braking for).
Physics 101. Actually, Physics 8.01. For the truly adventurous, check out Walter Lewin's physics videos on MIT's OCW, specifically the lecture dealing with friction. http://ocw.mit.edu/OcwWeb/Physics/8-01Physics-IFall1999/VideoLectures/detail/embed08.htm. This is roughly equivalent to an AP Physics class.
mray1031
03-01-2009, 20:44
zyck I think you and I are on the same train of thought. Still wondering about how to implement it though. Lets see if anyone else comes up with it.
Btw. Has anyone thought of just hanging out in the carpet and speeding around everyone? you'd have to be a great shot with those balls but it would be tough to score on you
gorrilla
03-01-2009, 20:48
Some quick suggestions off the top of my head:
Traction Control. This was mentioned before, but I'll summarize the idea of a traction control system. In a car, you compare the speed of the powered wheels to the unpowered wheels. If there is a difference, that means one set is slipping while the other isn't. The computer cuts a bit of power to the driven wheels until both spin at the same rate. For a 4-wheel vehicle, you can use an accelerometer to get the differential measurement.
Antilock Brakes. Antilock brakes don't necessary cut down on braking distance, but rather they give you control WHILE braking. If you slam on a car that doesn't have antilock brakes and you lock the wheels, the car basically keeps moving in the direction of its momentum vector, regardless of which direction the steering wheel is pointing (conservation of momentum). With antilock brakes, you're switching between static friction and dynamic friction, and in a nut shell, this gives you some control (ie allows you to swerve around the object you're braking for).
ive seen bakes before....would be tricky to get anti-lock brakes,although i dont think they would help that much, maybe if you had traction control and anti-lock brakes?
waialua359
03-01-2009, 20:49
All this talk about effectively driving in these low-friction conditions. I would instead think of how the other parts in your kit can help you score DESPITE the poor driving controllability conditions.
Two things off the bat I saw in the kit and my concrete memory of team 25 in 2006. The lazy susan turet (or equivalent) and the camera. ;)
Dracco1993
03-01-2009, 20:51
Yes, we thought about staying on the carpet but we decided it would be a bad idea because like he said, you would be an easy target and you will only be able to have at most 1/2 of your robot on the carpet.
P.S. How do you guy get thebluealliance name thing on there?
Zyck_titan
03-01-2009, 20:53
Mray
We thought of this, but our team decided not to do it Because;
A.) too much fabrication (time money etc)
B.) too much tuning to get it right (time)
But basically Thrust downward, with it you could dynamically control weight throughout the match, getting advantages of heavier btos and lighter bots, whenever you needed it.
Dracco1993
03-01-2009, 21:10
I dont think they will allow anti-lock brakes. They said at the kick-off that the only thing that can control the speed of the tires are the new jaguars and the old things. (I forgot what they are called.)
DarkFlame145
03-01-2009, 21:12
We cant use OMNI's or traction wheels, only the KOP wheels. any ideas of how to drive?
Very carefully
Dracco1993
03-01-2009, 21:20
I still need help on thebluealliance thing please.
Mike8519
03-01-2009, 21:22
You can pulse "brakes" electronically. Plus you can control the coast/brake setting on the jaguars with the Digital outputs
Alan Anderson
03-01-2009, 21:24
I dont think they will allow anti-lock brakes. They said at the kick-off that the only thing that can control the speed of the tires are the new jaguars and the old things. (I forgot what they are called.)
The "old things" are Victor 884 speed controllers from IFI.
Antilock brakes and traction control are implemented using software and sensors. The Victors or Jaguars are still used to control the motors.
gorrilla
03-01-2009, 21:25
I still need help on thebluealliance thing please.
try here
http://www.thebluealliance.net/tbatv/teambadge/teambadge.php
Zyck_titan
03-01-2009, 21:32
So here's my basic Idea for anyone who was following, Use a fan or such to provide thrust downward, or basically a (much) lower pressure area underneath the bot, the result (hopefully) would be more weight force downwards, making up for that lost weight force from the low traction, I'm not entirely sure exactly how to implement, and I'm not bothering to run the math on how much airflow is required to make up even say 200 lbs, but something like this (http://www.americanmuscle.com/mishimoto-dualfan-wshroud1.html) might be sufficient,
any thoughts?
gorrilla
03-01-2009, 21:38
So here's my basic Idea for anyone who was following, Use a fan or such to provide thrust downward, or basically a (much) lower pressure area underneath the bot, the result (hopefully) would be more weight force downwards, making up for that lost weight force from the low traction, I'm not entirely sure exactly how to implement, and I'm not bothering to run the math on how much airflow is required to make up even say 200 lbs, but something like this (http://www.americanmuscle.com/mishimoto-dualfan-wshroud1.html) might be sufficient,
any thoughts?
the rules specify that the Rover Wheels are the only thing allowed to provide "traction"
gorrilla
03-01-2009, 21:39
I still cant get it to be my signiture though. :confused:
after you do the colors, you have to get the "code" and put it in your signature....
Greg Marra
03-01-2009, 21:49
What if your robot had two drive trains? One in the forward direction and one in perpendicular to the forward direction. You lifted one drive train to use the other. Kinda like the trucks that are made to drive on the road and railroad tracks.
This is something that Buzz (FRC175) has done to great success in the past. Look up some old video - it's pretty spectacular.
pdelbarba
03-01-2009, 21:58
Isn't using a fan to create a low pressure zone not considered a traction device? I doesn't directly contact the ground nor does it actually constitute any solid form of traction as it would rely on a limited power supply. Such a device could not be used continuously throughout the match (just an estimate, no supporting math) as it would deplete the battery and rob power to the wheels. This is, of course, dependent on your drive configuration (four wheel/two wheel drive) and other systems such as pneumatics. This would, however, be very useful for acceleration and cornering if it could be designed with minimal spin-up time.
On a separate note, is there a rule stating that we cannot use any sort if thrust based propulsion system and just mounting the wheels as unpowered casters? Such a device wouldn't even rely on traction at all...
joeweber
03-01-2009, 22:11
How effective would it be to use an accelerometer and a rotation sensor to not allow the wheels to go far beyond wheel slippage. You could maximize the traction with out spinning all over the place.
Mike8519
03-01-2009, 22:20
Isn't using a fan to create a low pressure zone not considered a traction device? I doesn't directly contact the ground nor does it actually constitute any solid form of traction as it would rely on a limited power supply. Such a device could not be used continuously throughout the match (just an estimate, no supporting math) as it would deplete the battery and rob power to the wheels. This is, of course, dependent on your drive configuration (four wheel/two wheel drive) and other systems such as pneumatics. This would, however, be very useful for acceleration and cornering if it could be designed with minimal spin-up time.
On a separate note, is there a rule stating that we cannot use any sort if thrust based propulsion system and just mounting the wheels as unpowered casters? Such a device wouldn't even rely on traction at all...
Is that low pressure zone intended to increase your traction/performance on the playing surface? If it does, then I would believe it would be defined as a form of traction enhancement. Contact does not matter.
"fans" = shades of CanAm Chapparel 2Js. They only raced the one year way back... kinda funny to see shredded debris ejected by the two 45hp ducted fans though. Quickly banned as not fitting the FIA racing rules.
Homsar66
03-01-2009, 23:42
Is that low pressure zone intended to increase your traction/performance on the playing surface? If it does, then I would believe it would be defined as a form of traction enhancement. Contact does not matter.
Think. If this were true, then weight could be "a form of traction enhancement", similar to any traction control ect. Meaning anything on the robot is outlawed, as it adds weight to increase traction.
haha its like a car on ice...=D
NorviewsVeteran
04-01-2009, 00:40
haha its like a car on ice...=D
Obviously you don't drive the robot.
Akash Rastogi
04-01-2009, 00:46
Sorry to break it too you all but tank drive is going to be a pain, Tank drive has an axis of rotation around the center of the robot. BUT you have a trailer, you are going to be dragging the trailer wheels side ways it will be a battle.
Also the trailer anytime that you backup you will jack knife almost immediately, then you will be unable to control your backwards motion, turning with the trailer.
Hate to break it to you, but a lot of teams have some innovative programming planned. *hint hint* *wink wink*
There are always ways to work around obstacles...even if they're on the moon.
Akash Rastogi
04-01-2009, 00:49
The siguniture still does not work. =(
Guys no back to back conversations in threads please. Either stay on topic or use something like AIM or the PM function.
smartkid
04-01-2009, 00:53
^ No double posting, theres an edit button for that :P
pinkanese
04-01-2009, 00:55
So here's my basic Idea for anyone who was following, Use a fan or such to provide thrust downward, or basically a (much) lower pressure area underneath the bot, the result (hopefully) would be more weight force downwards, making up for that lost weight force from the low traction, I'm not entirely sure exactly how to implement, and I'm not bothering to run the math on how much airflow is required to make up even say 200 lbs, but something like this (http://www.americanmuscle.com/mishimoto-dualfan-wshroud1.html) might be sufficient,
any thoughts?
I have one, the fan in the link draws 8a of current and the hardware we have can only take 3/4a. Also I am thinking that there is little way of creating enough force to hold the robot to the Crater. One of my team mates did the math and estimated that you will need to move over 900 cubic feet of air per minute :). It might very well be possible to use a fan to create more downforce, I know it has been done in racing, but the question is going to be one of how much weight can you spare to add it in, how much room do you have for it, and more importantly can you spare that mush energy from your system to power it and all the other essential items.
Akash Rastogi
04-01-2009, 00:58
^ No double posting, theres an edit button for that :P
The post was meant to point out to specific people. ;)
Nevertheless, touche.
CraigHickman
04-01-2009, 00:59
Hate to break it to you, but a lot of teams have some innovative programming planned. *hint hint* *wink wink*
There are always ways to work around obstacles...even if they're on the moon.
Eh, figure it's time to stick my nose in here.
Anyway, Titan's idea about augmenting the robot's downward force is an interesting one. Keep working on that, you really could have something there.
Folks, FIRST just took the box we've all been thinking in and ripped it to pieces. Even a kit chassis is stuck in the high traction environment we've had. It's time to shed all these ideas of how we used to move, and really DESIGN around CONSTRAINTS, much the way the real world works. What you're all seeing and experiencing is a valuable skill, so don't blow this one off.
That being said, I have absolutely NO idea what to advise my team to do... I've spent all of today doing different designs, all of which end halfway when I realize different factors.
...
Has anyone felt the new playing surface? How much does it cost? Would plane old hallway concrete be a suitable substitute?
We bought two 4' x 8' sheets of that "ice" material ($30 per sheet) for the floor and did some testing. We compared traction on the "ice" the carpet and plain old tiled floor. The ice was the slickest. The carpet was slick but not as bad. The tile was surprisingly almost the same as the ice. So, to answer your question, if you cannot afford to buy some of the ice material, use a tiled floor as your best simulation.
gburlison
04-01-2009, 01:04
I have one, the fan in the link draws 8a of current and the hardware we have can only take 3/4a. Also I am thinking that there is little way of creating enough force to hold the robot to the Crater. One of my team mates did the math and estimated that you will need to move over 900 cubic feet of air per minute :). It might very well be possible to use a fan to create more downforce, I know it has been done in racing, but the question is going to be one of how much weight can you spare to add it in, how much room do you have for it, and more importantly can you spare that mush energy from your system to power it and all the other essential items. Many teams used 4 CIMS for their drivetrain in past competitions. It seems that 2 CIMS would be sufficient to drive the wheels this year. How much more traction would you get from 2 CIMS driving fans? what if the fans were used for purplusion instead? i.e. like an airboat.
gburlison
04-01-2009, 01:12
We bought two 4' x 8' sheets of that "ice" material ($30 per sheet) for the floor and did some testing. We compared traction on the "ice" the carpet and plain old tiled floor. The ice was the slickest. The carpet was slick but not as bad. The tile was surprisingly almost the same as the ice. So, to answer your question, if you cannot afford to buy some of the ice material, use a tiled floor as your best simulation.
Did you collect and numbers to help us compare?
I was thinking that we might stretch a large sheet of plastic over a tile floor and tape down the edges. If the plastic is thick enough, hopefully the robot will not wrinkle and tear it long enough to practice.
Akash Rastogi
04-01-2009, 01:15
Many teams used 4 CIMS for their drivetrain in past competitions. It seems that 2 CIMS would be sufficient to drive the wheels this year. How much more traction would you get from 2 CIMS driving fans? what if the fans were used for purplusion instead? i.e. like an airboat.
hahaha, I do love these propulsion ideas.
But i think that teams are making this much tougher for themselves by not looking at what by some people has been described and explained to me as simple physics. Why make it harder on yourself by making that hypothetical propulsion robot? Think of the new challenges you create for yourself. A part of engineering, as what I've been taught by my grandfather, is not only creating solutions, but moving closer to the obstacles at hand, and not creating new ones for yourself.
Sure propulsion would be innovative in an FRC robot, sure it'd be tons of fun, but look at the new challenges you're bringing up for your own team. How would this effect the physics of you carrying the trailer? When you turn, the inertia of that trailer will lead you into a corner with propulsion. A hybrid of propulsion and wheels is almost useless even if it is deemed allowed (which by the "no other traction on the field allowed" part of the rules can't be done anyway).
Ranting aside, my point is just for teams to think more clearly in their ideas. I'm not an engineer so I'm not saying you even have to listen to me or consider my advice, just don't create new obstacles for yourselves.
So does anyone want to chime in about using these wheels as omniwheels. Would it even be possible. I am feeling slightly lost since all of my preseason designs involved mega-traction and multi-speed high torque transmissions
Yup Sean, it looks possible. I've gone over the CoF with a lot of people and depending on the orientation of your wheels, you could in theory do a Kiwi type holonomic drive. I'd rather test it out though.
sdcantrell56
04-01-2009, 01:16
So does anyone want to chime in about using these wheels as omniwheels. Would it even be possible. I am feeling slightly lost since all of my preseason designs involved mega-traction and multi-speed high torque transmissions
PowerOfKings
04-01-2009, 02:59
So Zyck_titan it seems you had the same idea I did... >.< and I thought I was original.
We did the math and with a 480 watt (the maximum wattage motor allowable) you could produce just under .2 psi. (allow variance for rotor blade type) Numbers for this were taken from a website publishing the Static Pressure capabilities of their 480 watt fans. (we then did conversions to psi)
Now you have to multiply by the square inches (maximum allowable 1064 = 28x38) and we get about 213 pounds of extra force.
So if you had a skirt around the entirety of your robot you could conservativly expect to roughly double the weight/normal force of your robot. (you'd lose a lot due to inefficiencies)
This means that when applied to F=MA in regards to your acceleration you can expect to take it from roughly .6 m/s (static friction with no fans) to 1.2-1.5 m/s (static friction with fans)
Now logistics might be a tad tough but hey if it gives you double your friction then it's worth it in my book. The real problem... ripping up the play area. If it isn't secured you would just vacuum it up and be completely stuck. This seems a real problem to me since when you look here ( http://thebluealliance.tumblr.com/post/68204495/hands-on-with-lunacy-field-elements ) at the end you can see the edge of the field which is merely secured with TAPE and now if this is a thin linoleum style roll you would just suck it right up and then... you wouldn't be able to move and might even get a penalty for destroying the field.
...after further research ( http://www.frpshop.com/pdf/installation_instructions.pdf ) assuming FIRST followed these rules for the construction of a gasolenar frp floor it might not be too much of a problem...
dtengineering
04-01-2009, 03:23
So does anyone want to chime in about using these wheels as omniwheels. Would it even be possible. I am feeling slightly lost since all of my preseason designs involved mega-traction and multi-speed high torque transmissions
Omniwheels are specifically designed to have a lower coefficient of friction in the transverse direction than in the inline direction.
These wheels, according to the manual, have a transverse coefficient of friction that is about twice that of the inline direction. They are, by my reading... anti-omni wheels.
Jason
As I think more and more about this game I've come to a conclusion.
No matter what you do, you'll still be working with limited traction. Instead of worry about getting more traction, why not just embrace the lose of traction?
I think the question of "how can we drive on this new floor?" should be posed to the drivers. It'll require quite a bit of finesse and skill to master the lack of traction but those who do will be at a huge advantage.
Perhaps you should have your drivers research things like Drifting, ice driving, and driving in the rain to get some ideas. I personally, dabble in a bit of drifting myself so I'd like to see if I can apply that experience to driving a robot on this surface.
I agree
The coefficient of transverse static (or kinetic for that matter) friction is significantly higher than the Inline coefficients of friction. I can't help but imagine that some use could be had from the transverse friction.
The the higher friction could be used to create a centripetal force to assist in turning, which would require that the wheels rotate.
Or, a foot could be made from a perpendicularly positioned wheel, and use the transverse coefficient of friction to help stop the robot. Would that be legal?
PowerOfKings
04-01-2009, 04:54
It doesn't make sense to me why one would be higher than the other the wheels are molded and don't have any sort of "stepping" that would cause this ... any idea as to why one would be greater than the other?
GaryVoshol
04-01-2009, 06:53
We bought two 4' x 8' sheets of that "ice" material ($30 per sheet) for the floor and did some testing. We compared traction on the "ice" the carpet and plain old tiled floor. The ice was the slickest. The carpet was slick but not as bad. The tile was surprisingly almost the same as the ice. So, to answer your question, if you cannot afford to buy some of the ice material, use a tiled floor as your best simulation.
Did you do any testing on terrazzo floors? (The stuff where a slurry is poured, and then polished after it hardens)
Alot of people's response to this problem reminds me of an exercise we had to do at a workshop that work had us do to advocate thinking outside of the box. They had us put up our hand and the guy would push it. Of course out of reflex we would resist. And he asked us why we resisted? Because it our habit to do so. Sometimes instead of just resiting the forces that are working against you is to loosen your control and work with those forces and allow them to do the steering.
Seems to me what people have the biggest problem here with is being in complete control of their robot and are trying to come up with every possible solution to regain that control when maybe the best solution is to embrace the lack of control and use it to the advantage like the guys who suggest the drifting training. You could waste alot of time and energy otherwise.
Dracco1993
04-01-2009, 08:06
Some of my team members and I calculated that using 2 sim's and suction cup properly you caould add an extra 600 pounds to your robot with ever touching the ground... Any Ideas?
Some of my team members and I calculated that using 2 sim's and suction cup properly you caould add an extra 600 pounds to your robot with ever touching the ground... Any Ideas?
Shade of 2003.
Teams used suction cups to plant themselves at the top of the ramp. That worked for the first few matches and then after getting ran over by robots for a while the field surface got worn down a bit and the suction cups stopped working.
I suspect the same would happen this year as well.
Mike8519
04-01-2009, 08:25
Think. If this were true, then weight could be "a form of traction enhancement", similar to any traction control ect. Meaning anything on the robot is outlawed, as it adds weight to increase traction.
Yes, weight does help traction, but it is a legal addition to the robot to a limit. Increasing your traction beyond your idle state/curb weight I think would cause the issue.
Traction control is legal because it is increasing control electronically. Not changing the physical interaction between the playing surface and the rover wheels
Dracco1993
04-01-2009, 08:27
Shade of 2003.
Teams used suction cups to plant themselves at the top of the ramp. That worked for the first few matches and then after getting ran over by robots for a while the field surface got worn down a bit and the suction cups stopped working.
I suspect the same would happen this year as well.
I see your point, but that math was done assuming that the suction cups are about 1/2 to 1/4 of an inch off the ground. They would be more like a hover craft in reverse then suction cups.
A hybrid of propulsion and wheels is almost useless even if it is deemed allowed (which by the "no other traction on the field allowed" part of the rules can't be done anyway).
I don't think it is useless. The maximum force that can be generated with the wheels for a max-weight robot is 26.7N. With some math, you can determine that 5N and 10N fans could be plausibly added to your robot, which would add 18% and 36% to the acceleration of a max-weight robot, respectively. Some of that would be eaten by the airstream hitting the trailer, but I think a fan approach could actually increase performance.
I think there is a reading of the rules where it would be allowed. Traction generally refers to the contact between a wheel and the ground. Fans aren't enhancing your traction with the field (and unless you have a quickly reversible fan, they actually hurt your braking performance), they are applying a force by pushing against the atmosphere. However, there are also readings of the rules where it wouldn't be allowed, and it'll have to wait until QA opens to see if they are allowable.
Dracco1993
04-01-2009, 08:36
Last year we were one of a few teams that used a vacume to pick up the ball so we have some experience with them. One of my friends on the team picked up the ball just using his mouth. He sucked on a vacume tube atached to a cake holder lid. With that much of a surface area he could pick up the ball for about 5 seconds.
Does this make any sense? Six wheeled bot. Four corner wheels driving forward and two side centered wheels driving backward as needed. Or, one wheel dead center of bot that would spin oppisite as needed. Then hand the whole thing over to programers. They can make it work.
Look, all the ideas with fans:
Why not take your 200 pounds of force going downwards, point that to a side, and use it as a means of propulsion? It would certainly be a lot more efficient than using the force to increase the traction (it will still be low traction).
-Vivek
spiffyspleen
04-01-2009, 11:09
Would a holonomic drive system work well? I think it would make it easier to drive, but I don't know for sure. Here is a link to a picture of what I am talking about except that we would have to use the required wheels instead of the omnis. The required wheels should work as well though, because they barely have any traction.
http://wiki.chiefdelphi.com/media/0/09/2047_holonomic.jpg
smartkid
04-01-2009, 11:43
You would have to use the wheels in the kit as the casters, the wheel would be huge and bulky but it could work. Now you have to know that an omni-drive works on the principle of wheel slippage. It is designed so that wheel casters slip.
While this might make sense initially, holonomic actually starts looking very bad after viewing the COFs for the wheels.
10.2.4.1 Wheels
The wheels supplied in the 2009 KOP are very different from previous years’ kit wheels. The
tread material is Celcon M90, and has the following coefficients of friction on white, rippled
fiberglass plastic sheet
Inline, static: 0.06
Inline, dynamic: 0.05
Transverse, static: 0.14
Transverse, dynamic: 0.10
Yes, the transverse COF is very low, but the inline COF is less than half of it! This means a holonomic drive robot (with the COG in the middle) theoretically wouldn't move at all! In practice, it might, but not much.
You are absolutely right. Drivers will need to train themselves and adapt to the slick situation. Additionally, I believe that programming will also have a big part in this years game (i.e. Traction Control). Our team will be doing a lot of testing in terms of the best method of acceleration for our robot, inputting that into our system and applying that to our controls.
That could be a problem. In general PID controllers have always been a robust and great way to move your robot. The problem is that the ideal way to control the motors is with some type of ramp input meaning that the integrator may or may not work as exactly the same.
cj.reeves
04-01-2009, 14:05
is the coefficient of friction independent of surface area or not?
i think more wheels equals more friction
Zyck_titan
04-01-2009, 15:17
the rules specify that the Rover Wheels are the only thing allowed to provide "traction"
But the Fan is not providing traction, it's providing a weight force, making up for the lost 5/6 of weight
sbrumund
04-01-2009, 15:23
In simple physics terms friction is independant of the surface area.
In reality because we are on a textured surface it may not be.
If the wheels are the same as last years kit wheels the surface is a soft compound. With sufficient weight on each wheel the soft compound will conform the the bumpy surface producing more friction than a wheel riding on the top of the bumps.
The question then is what is then optimal load on a wheel to obtain the most friction. A simple experiment can be designed to determine the optimal loading and therefor the optimal number of wheels.
If allowed the vacuume design is possible. Many schools physics departments have a 4' diameter hovercraft where the lift is achieved with a battery powered leaf blower. The leaf blowers run well for 10 minutes on a charge and can easily support a 250 lb load. I believe the battery supplied has sufficient energy to power the leaf blower and drive motors and increase the normal force and friction. You would need a fully charged battery for each match.
Since we are simulating 1/6 gravity on the moon and there is no atmosphere on the moon I suspect a vacuume design wil not be allowed.
It is possible to get used to driving on a low friction surface. Those who are used to driving on ice and snow understand the limitations. Keep in mind that when driving a car on ice you are in the car and have a, seat of your pants, feeing of what is going. You do not have this sensor input at a remote driving station.
It is possible and easy through programming to limit how fast you can accelerate and turn. It takes some time to get used to, like the first time you slam on the brakes of a car with anti lock brakes.
Creator Mat
04-01-2009, 15:40
What if we did a air boat type drive train. where we take a floor fan gut the motor and replace it with a shaft and a belt drive to a few CIMS (or direct connect). Then our wheels would be attached to casters. To steer we would have directional flaps powered by servos on the back.
+: could defend the trailer if powerful enough, takes away some traction problem.
-: stopping could be an issue (but thats what brakes are for), and the power of the fan could not be powerful enough to move the robot.
adaptingthought
04-01-2009, 15:43
Im with the fan idea for propulsion. Not sure if it's legal though
cj.reeves
04-01-2009, 15:53
You want as little interaction between the wheels and floor as possible, although the fan is risky there are other ways to move your robot without interaction with the sourounding medium. As we all know a tank steer or skid steer is out of the question a 40lb flywheel spinning at a couple thosand rpm spun by the windsheild wiper motor on the horizontal axis should spin the robot on a dime just how NASA stabalizes their satelites in space it works on SW 2008
Creator Mat
04-01-2009, 16:04
Im with the fan idea for propulsion. Not sure if it's legal though
How would the fan not be legal? It could stay within size constraints, you would have to enclose it but that wouldn't be that hard if you use a premade fan, and you still use the wheels on the ground.
Creator Mat
04-01-2009, 16:07
You want as little interaction between the wheels and floor as possible, although the fan is risky there are other ways to move your robot without interaction with the sourounding medium. As we all know a tank steer or skid steer is out of the question a 40lb flywheel spinning at a couple thosand rpm spun by the windsheild wiper motor on the horizontal axis should spin the robot on a dime just how NASA stabalizes their satelites in space it works on SW 2008
But how would you move across the field? the flywheel would make you spin but the trailer would mess that up, plus to put in a fly wheel would take up a lot of space.
Can the battery power a large fan for 2:15' straight???
adaptingthought
04-01-2009, 16:51
someone found a fan thats legal by FIRST regulations
http://www.modelflight.com.au/rc_model_accessories/wemotec_micro_fan.htm
i dont know the calculations to find out how long the battery could power it though
Creator Mat
04-01-2009, 16:57
don't know might not but you only use the fan as the major draw of power off the battery i don't see why not. think about it using 4 CIMS to move a robot on wheels with high friction/wieght and a pnuematic system drained most of our battery last year. but if we where to use only the fan it would be 1-4 CIMs moving at most a 5 pound fan (im being very heavy) in a circle constantly, in my opion would not drain the battery that much.
Creator Mat
04-01-2009, 17:00
someone found a fan thats legal by FIRST regulations
http://www.modelflight.com.au/rc_model_accessories/wemotec_micro_fan.htm
i dont know the calculations to find out how long the battery could power it though
im not using any rules as backing here but from what i know as long as the fan is enclosed (no sharp/dangous appendages) and you use CIMs to power it not the motor that came with the fan (illegal motor) the fan would be legal
Does this make any sense? Six wheeled bot. Four corner wheels driving forward and two side centered wheels driving backward as needed. Or, one wheel dead center of bot that would spin oppisite as needed. Then hand the whole thing over to programers. They can make it work.
I would guess this is roughly the setup you had in mind with wheels:
| |
_ _
| |
(with possibly more series of sideways and forward-facing wheels)
Vertical wheels are forward facing, horizontal are sideways facing, in case people don't know.
And that wouldn't exactly be difficult to program, but the operators would have to have incredible coordination, since assuming you're driving in tank mode it would require three joysticks (two for each set of wheels, plus one more to operate the side wheels.) Add more motors if you want additional sets of wheels, unless you can put three wheels on one motor...
Either that or have the side wheels spin in conjunction to the outputs of the joysticks (i.e. if left is spinning full forward and right is spinning full backward, the wheels will turn in the direction that will provide the most traction. I'd have to run a simulation to determine exactly which would be the best way, whether you want more turning force, causing the trailer to possibly fishtail, or less to stabilize it.)
Again, not apparently difficult to program.
edit: (with random concerns I think of)
The only thing I could see right now is collisions of of the chains (if they're used to provide the moving force for the wheels) from the forward/sideways wheels. Seems it could be remedied by using varying radii with the grabbers on the wheels (spaced the name of them) that move the chain.
PowerOfKings
04-01-2009, 17:21
Look, all the ideas with fans:
Why not take your 200 pounds of force going downwards, point that to a side, and use it as a means of propulsion? It would certainly be a lot more efficient than using the force to increase the traction (it will still be low traction).
-Vivek
The 200 pounds of force comes from the fact that it is a vacuum creating pressure underneath the robot so in a 10x10 area with 1 psi (pound per square inch) you would have 100 pounds of force. You would not have nearly the same force from simply the fan on the side.
However we calculated that it would only take 7 pounds of force applied from the side to start a robot moving sliding sooo...
Cyberphil
04-01-2009, 17:48
Did anyone else see the size of that fan? Do you really think that would be worth buying 30 of those for a little bit more propulsion? Personally, I think its a waste of time, and how can we have tank treads if the rules state: "The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub). No other forms of traction devices (wheels, tracks, legs, or other devices intended to provide traction) are permitted."
someone found a fan thats legal by FIRST regulations
http://www.modelflight.com.au/rc_model_accessories/wemotec_micro_fan.htm
i dont know the calculations to find out how long the battery could power it though
Unless I'm off my rocker, 200g of thrust means a little less than a half-pound of thrust? You'd need something like 14 of them to be roughly equivalent to the amount of force produced by frictional interaction from wheels.
I made that approximation based off a calculation of maximum frictional force by multiplying the weight (120lb.) by the *kinetic* coefficient of friction (.05 from the first manual), yielding a maximum force of 6lbs. to accelerate the robot. So, that's a lot of fans. Running all of them at their maximum of around 300 Watts each, that accounts for 4200 watts, and, I could be wrong, but I believe that would mean it would discharge one of the 12v 18Ah batteries in around 3 minutes.
So, that particular fan sounds pretty in-efficient to me. I would suggest looking for one with a much larger prop. Larger, slow moving props tend to have a higher efficiency (not to mention they tend to be a little safer). A larger prop would also mean more thrust, and fewer fans. Knock the number of fans needed down to 4 or so, and it would be manageable I think.
I'm still intrigued by the possibility of a fan-propelled bot. I think it would be quite spectacular to watch in action.
*EDIT*
This is all completely without regard to the legality of the fan and motor, merely assessing the viability of the particular product.
Did anyone else see the size of that fan? Do you really think that would be worth buying 30 of those for a little bit more propulsion? Personally, I think its a waste of time, and how can we have tank treads if the rules state: "The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub). No other forms of traction devices (wheels, tracks, legs, or other devices intended to provide traction) are permitted."
I think they were referring to what is called "tank drive" which you can think of like driving a bulldozer, only without the tracks, just the wheels. The wheels on one side would operate together independently from the other side.
chuckmerja
05-01-2009, 12:47
Interesting that game team used "traction" with respect to wheels etc, and didn't use the word "propulsion", thus maybe leaving a fan solution open to possibility. Hmm
I think one thing that will be incredibly important is weight distribution, if you have more weight on one side your robot it going to be spinning in circles since that side will have more traction.
Cyberphil
05-01-2009, 15:31
I think they were referring to what is called "tank drive" which you can think of like driving a bulldozer, only without the tracks, just the wheels. The wheels on one side would operate together independently from the other side.
No. Sorry, but someone was talking about putting treads on their bot. Just clarifying that.
Akash Rastogi
05-01-2009, 15:33
No. Sorry, but someone was talking about putting treads on their bot. Just clarifying that.
Yeah I saw that somewhere...
People need to learn to use proper terms though please. There's listings on here about all different common terms used in FIRST and on CD, just search them people.
No. Sorry, but someone was talking about putting treads on their bot. Just clarifying that.
ahh, I apologize, in my haste reading through the threads I must have mis-read something :rolleyes:
In that, I would most definitely agree with you :)
Has anyone thought about creating a way to modify your center of gravity/weight distribution on the wheels, in order to aid in traction control? Would require a mass on board able to be moved in some direction or another. Might not be likely to have a light enough bot in most cases for it to be feasible. I wonder to what extent something like that would be able to help though? Hmm... possibilities...
nathanww
05-01-2009, 19:31
Our team was considering this, but then decided to go with automative-type traction control just on the basis of it being a proven technology
Creator Mat
05-01-2009, 19:44
a problem i have in general with all drive trains that allow u to turn suddenly like crab drive, swerve drive, and holonomic especally. with the all ready low traction of the feild a quick turn would send your robot flying down the feild out of control. then if you solved the traction problem your trailer would jack knife and take you out. also for the holonomic drive this drive in a high friction envrioment had no pushing power. so this year the holonomic drives biggest strength, its mobility, is gone. also the drive itself is made to have no pushing power and in my opion this year the robot with the most pushing power is king not the one with the most traction.
pacoliketaco
05-01-2009, 19:51
So today i went with my mom to buy a sheet of the flooring material. As has been stated before, it was $30 for a 4'x8'x.125" sheet, of a material that seemed extremely close to the look and feel of what people have already described. As soon as we got home, i took one of the new wheels and slid it on the flooring. At first, when we first opened the kit, i took a wheel and slid it on a table nearby, remarking at how there was no traction with the table surface. Well, compared to the actual flooring, that table would be much better. There is soooo little friction between the wheels and the flooring. i am actually quite scared by how little traction we are going to be working with. earlier someone was talking about these wheels being similar to those of previous years, in that they might deform a little to adjust to the uneven surface of the floor. That is not going to happen, as these wheels are very hard plastic, which will not be changing it's shape ever.
Not an expert on this but I think you should be limited to using 4 wheels. If not
then you could use 8 - 16 whatever - a wheels arms race.
If everyone has to work with four wheels the mental arms race has to be where the competition goes.
I think with the mass of the trailer behind you - best movement strategy is based more on sailing/sledding/bumper cars then driving on carpet.
sdcantrell56
05-01-2009, 20:16
Not an expert on this but I think you should be limited to using 4 wheels. If not
then you could use 8 - 16 whatever - a wheels arms race.
Adding more wheels does not increase traction. The maximum force you can apply to the floor is determined by the coefficient of friction and the normal force. Surface area has no influence.
Not an expert on this but I think you should be limited to using 4 wheels. If not
then you could use 8 - 16 whatever - a wheels arms race.
If everyone has to work with four wheels the mental arms race has to be where the competition goes.
I think with the mass of the trailer behind you - best movement strategy is based more on sailing/sledding/bumper cars then driving on carpet.
Adding more wheels would probably make little difference in driving, and would probably actually make it harder to turn (if you use skid turning, and with that many wheels, you probably will). Also, in theory adding wheels won't really help since as long as you're using the same wheels (you will be), you're just spreading out your normal force while still having the same coefficient of friction (.06).
nexgendedhed
05-01-2009, 21:18
I am not a physics expert either, but do have one thing that might be useful.
People are using the terms traction and friction almost interchangeably, which may lead to misconceptions. Friction (F=uN) is a fairly simplistic empirical model that does not depend on surface area. Traction, on the other hand, actually does depend on surface area. Using a larger surface area while leaving other factors constants results in increased traction. For evidence, think of rubbing two small pieces of sandpaper together versus two larger pieces, or wide drag racing tires.
I can't find a whole lot of information on traction, however, so if anyone knows more and would like to correct me, I would appreciate it.
sxysweed
05-01-2009, 21:35
I stopped reading the thread at page, two it's nearing bed time for me. But I found the solution for those having a hard time TURNING IN PLACE with Skid Steer. Turning in place in difficult enough, but remember the TRAILOR. Not only do you have to pull the wheels sideways, but you have to pull the skid sideways. Trick for the drivers would be to drive forward and then turn. Initiate a drift and then use that to slide. The turns would be wider, but more controlled. It would take practice though.
Example of what I mean:
http://www.youtube.com/watch?v=P5aWAW2hea4
But another proble I see happening is a few teams that have no practice or experience driving with no traction will go straight forward and before even attempting to stop, they will try to stop and either lock of the wheels or put them in reverse. You will stop like this at some point, but I find it likely that the robot will veer off to one side. These are things I've experienced in trying to drift RWD RC cars with PVC Tires. It's different with Skid Steer, but the lack of traction idea is similar.
Adding more wheels does not increase traction. The maximum force you can apply to the floor is determined by the coefficient of friction and the normal force. Surface area has no influence.
Are you sure?
I'm more involved in 1:1 cars and racing, and the general idea is that wider tire = more traction. Same reasoning behind the Bugatti Veyron's 16" wide rear tires.
But to me it would seem that having two/three wheels directly next to each other would make turning more difficult, as the mu is greater sideways than forward. And this is from my observed experience in the FLL....
Another thing I noticed today is that the Rover Wheels aren't true. What I mean by this is that they're completely uneven all the way around. Try rolling one forward, it will fall over to one side real quick. I'm sure mounting them normally and giving them a good once over on a good surface (concrete?) will level it out nice, and prove more surface area. But acording to the guy in quotes, this would be useless?
-----------
Me and my AP Physics teacher had brief discussion on using fans or duct fans (I would prefer the later for a few other reasons) to aid in propulsion. You could possibly add another 10N for acceleration, but you also forget you can possibly use this to steer the robot. Mount a few small duct fans here and there that generate say 3N each, and you can help correct a drift or even slow down in a straight line....
=Martin=Taylor=
05-01-2009, 21:56
We got a 4 wheeler running on the game floor today with the wheels.
Seemed to be controlable enough. We could turn it when it weighed 30 lbs. and when it weighed 120~160 lbs. it turned even better.
We'll have to see how it goes with the trailer.
We got a 4 wheeler running on the game floor today with the wheels.
Seemed to be controlable enough. We could turn it when it weighed 30 lbs. and when it weighed 120~160 lbs. it turned even better.
We'll have to see how it goes with the trailer.
that's cool to hear. how is it set up? tank-drive?
RoboGeek99
05-01-2009, 22:53
Ok weve spent all of our meeting today discussing the "best" type of drivetrain...we narrowed it down to a tank style and a car steering with back-wheel drive...any thoughts?
Ive read the thread and i really cant see much discussion on the drive trains but rather on increassing friction...my advice embrace the challenge instead of trying to move around it
PS-Wat is the aprox weight of the trailer rules dont specify?
samir13k
05-01-2009, 23:17
I replaced the wheels on a 30-40ish lb kitbot today that was previously made with the new kitbot wheels, and tried it out on the "regolith" surface, and so far....its not looking so well. The bot can be kicked around and will slide sideways really easily. Of course, there are many different factors going into play, especially the low mass.
=Martin=Taylor=
05-01-2009, 23:21
that's cool to hear. how is it set up? tank-drive?
"Long" drive configuration with a wheel at each corner. Normal tank-steering.
Geared for around 8 fps.
I replaced the wheels on a 30-40ish lb kitbot today that was previously made with the new kitbot wheels, and tried it out on the "regolith" surface, and so far....its not looking so well. The bot can be kicked around and will slide sideways really easily. Of course, there are many different factors going into play, especially the low mass.
Weight makes a HUGE difference.
Also realize that other bots won't be able to push you as hard as you can kick.
Crashes are a different story.
Jbrown1011AZ
05-01-2009, 23:45
Do you give any hints to people who have no experince with driving in the snow oricy conditions
pfreivald
06-01-2009, 10:55
I live in the land of ice and snow, and tips for driving in these conditions are relatively self-explanatory, governed by this rule:
If you lose static friction, you lose control.
Driving a *car* in such conditions means that you:
A. Go easy on the accelerator to avoid slipping.
B. Go easy on the brake, for the same reason.
C. Turn gradually, slowing down much more than you normally would.
D. If you start to skid, turn into the direction of the skid to allow your wheels a chance to catch again.
Now, how this applies to different drive train configurations in Lunacy is another matter!
Patrick
Ben Bastedo
06-01-2009, 11:03
I live in the land of ice and snow, and tips for driving in these conditions are relatively self-explanatory, governed by this rule:
If you lose static friction, you lose control.
Driving a *car* in such conditions means that you:
A. Go easy on the accelerator to avoid slipping.
B. Go easy on the brake, for the same reason.
C. Turn gradually, slowing down much more than you normally would.
D. If you start to skid, turn into the direction of the skid to allow your wheels a chance to catch again.
Now, how this applies to different drive train configurations in Lunacy is another matter!
PatrickThanks for the tips Patrick. Don't get too much of that weather down here in Texas.
Ben
pfreivald
06-01-2009, 11:18
No worries.
On another note, I think it would be a huge mistake to try to design a robot that is going to maintain traction all the time -- because it won't, especially when it gets hit by high-speed *other* robots. It would be much better to *know* that you are going to slide, and then try to figure out how you can use this to your advantage in controlling your bot. Driver practice will be paramount.
Someone suggested finding a video game (I think it was MarioKart(?), although I am sure there are others) that involve driving on slippery surfaces, just to get used to the general idea of how things handle.
Patrick
Ben Bastedo
06-01-2009, 11:24
HA, basing how a robot drives off a video game! That is new, but an awesome idea! :)
pfreivald
06-01-2009, 11:38
You could easily go too far with that idea, methinks, but it's not a bad way to get those of you who reside in warmer climates more accustomed to this kind of driving.
Patrick
Enigma's puzzle
06-01-2009, 12:00
Your Normal Force is a set number, however much your robot mass. That is divided by the surface area that is touching the ground "Theoretically" (It actually depends on Weight Distribution too). The Mass that is supported by each wheel can actually be measured by load cells (And then divide by the surface touching the surface area making contact with the ground to find pounds per square inch.)
Adding or subtracting wheels (if all are powered) doesn't actually make the difference in total, It changes the pounds per square inch of each wheel but not the traction because your total mass is still the same and the friction is still the same.
However if all wheels are not powered then you lose traction because whenever you try to drive not all of your mass is being used to effect friction.
SO more wheels doesn't really help except for the stability of the robots. you won't gain more traction from more wheels.
The hieght of the trailer (the posts around the outside) goes linearly from 42 to 34 inches, I am not sure if that is from the floor or the base of the trailer.
Elgin Clock
06-01-2009, 12:03
Use springs. ;)
Matthew2c4u
06-01-2009, 12:14
i still see alot of constrained thinking about the drive train,Thinking outside the box has helped us the most.
Also, USB Joystick, not last years joystick.
But what about momentum? Changing directions with a heavier bot it's going be be a bit harder I think... Or maybe not, I'd like to get some input on this.
it is because "whats in motion tends to stay in motion" physics 101
Enigma's puzzle
06-01-2009, 14:26
Mario cart, SHERBET LAND!
The biggest tip to people not used to icy condition, is Don't break traction, accelerate slower than you think you need to, because if you try to accelerate too fast you will spin and lose traction, causing you to have to slow down your wheels to recover. essentially making you accelerate slower.
Bob Steele
06-01-2009, 16:52
Just a small note from me...
I believe that one has to think of this game more in terms of thrust vectoring rather than driving.
These robots will behave much more like spacecraft than they will cars.
In your brainstorming... think about applying force in the direction you want the robot to change its acceleration.
Earlier another post showed that the maximum acceleration (assuming a 120 pound robot and .o6 coefficient of friction) that could be achieved is around .588 m/s2...
Apply that thrust in the direction you want to change the robots acceleration.
This would seem to require that powered wheels (and all wheels need to be powered to get maximum traction) are capable of changing their vectors in 360 degrees. This is because the only thrust that the robot is capable of is from these wheels...
The only other thrust would come from those same wheels turned sideways..and that only would be when the robot is moving and it would be in the negative direction to the motion. This would bring the higher transverse coefficient of friction into play and could be a very good way to de-cellerate the robot. It could conceivably de-cellarate the robot with nearly twice the magnitude as the powered wheels...
just something to think about.
don't think driving... think piloting...and think piloting of a space craft..
thanks
good luck at the competition!!!
May the Net Force be with you!!!
hi think slow for now we can predict all we want but the real test is going 2 come at regional, the ones that drive the best will do the best in my opinion. my idea learn fast.
Eaglepilot52
06-01-2009, 19:47
How to drive on the new arena surface? Well, stated in the rules, ALL robots have the exact same wheels. They cannot be modified. So I guess to drive well, you need a good driver, a good robot design, and some luck.
However, I heard that having less wheels would be better. I think mabye it's becuse there's less traction or something like that...
"Long" drive configuration with a wheel at each corner. Normal tank-steering.
Geared for around 8 fps.
Weight makes a HUGE difference.
Also realize that other bots won't be able to push you as hard as you can kick.
Crashes are a different story.
Were the wheels maintaining a 'static' state? or were they spinning a lot?
Hehe, sorry for asking so many questions. I'm interested in the first-hand experience. My team hasn't gotten their hands on the regolith flooring yet, and are considering a 4-wheel tank drive. we're considering a wide frontal orientation for the wheels though. so the front of the bot would be the longest side. we'll see if this helps improve turning ability. I think we're going to be up for a LOT of testing (trial and error) this year. haha.
ezygmont708
06-01-2009, 22:13
We have a tank drive that we use each year for prototyping. Please take a look at the video on the link below...
http://www.youtube.com/watch?v=s1Rl8mcTpjw
This is just last year's code running two CIMs per toughbox. Just a little video taste....
swamp_child
06-01-2009, 22:31
A holonomic drivebase would still be possible even with just the KOP wheels. The coefficient of friction is low enough that a swerve base or kiwi drive would both be possible, since the wheels slide easily enough.
Kiwi doesnt quite work, because the coefficient of friction when moving laterally is far greater than that of when the wheel is moving normally.
RoboGeek99
06-01-2009, 23:04
We have a tank drive that we use each year for prototyping. Please take a look at the video on the link below...
http://www.youtube.com/watch?v=s1Rl8mcTpjw
This is just last year's code running two CIMs per toughbox. Just a little video taste....
What looks like could be a problem with that is the trailer is going to jacknife and that could make a huge diference
pacoliketaco
06-01-2009, 23:16
although my team has already bought some of the flooring, we have yet to test it out. hopefully soon we will be able to retrofit an old robot to make use of the new wheels. i think that practice is really going to make a difference this year.
on a separate note, ive read a lot of posts about driving this year, but no one yet (as i have seen) has brought this up:
from rule regarding wheels (i dont remember the number)
...a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub)...
if you ask me, we dont have to have the wheels spinning in a circular orbit. by building a hub that allows the rotation of the wheel to be offset from the center, a different driving stlye would result. im not sure if this would help at all, or be legal for that matter, it just seemed interesting.
=Martin=Taylor=
06-01-2009, 23:53
Were the wheels maintaining a 'static' state? or were they spinning a lot?
Hehe, sorry for asking so many questions. I'm interested in the first-hand experience. My team hasn't gotten their hands on the regolith flooring yet, and are considering a 4-wheel tank drive. we're considering a wide frontal orientation for the wheels though. so the front of the bot would be the longest side. we'll see if this helps improve turning ability. I think we're going to be up for a LOT of testing (trial and error) this year. haha.
It didn't drive so well once we got the trailer going today :(
Our driver did develop a clever method of jack-knifing the bot back and forth to turn around. Simply tank steering was impossible with the trailer...
We're looking into the wide config also... Let me know how it goes for you.
Enigma's puzzle
07-01-2009, 08:57
We're looking into the wide config also... Let me know how it goes for you.
<R18> part b
The Trailer Hitch must be rigidly attached to a fixed location on the ROBOT, with the long dimension of the Trailer Hitch horizontal and the opening of the C-channel facing away from the ROBOT. The horizontal center line of the Trailer Hitch must be 2-13/16 inches above
the floor.
this may inhibit wide robots but im not sure?
chuckmerja
07-01-2009, 09:11
Team 1696 "driving" last year's bot with new wheels on FRP - ufduh!
http://www.youtube.com/watch?v=Zc_L3Cbz5nQ
http://www.youtube.com/watch?v=WbG5Ft72z9I
dudefromjapan
07-01-2009, 11:02
We got a 4 wheeler running on the game floor today with the wheels.
Seemed to be controlable enough. We could turn it when it weighed 30 lbs. and when it weighed 120~160 lbs. it turned even better.
We'll have to see how it goes with the trailer.
How's the acceleration and breaking on it?
Boco Gear
07-01-2009, 12:03
Wow there are a lot of ideas being thrown out here. I have to see in all of them games I have seen for FRC this is definitely something new. As for driving on this new surface it seems hard.
JUst wondering wouldn't the drifting ideas and driving on ice ideas kind of be hindered from the trailer a bit?
Tom Line
07-01-2009, 13:59
Does anyone have any idea why WildStangs removed their driving video from youtube?
Andrew Bates
07-01-2009, 14:29
Probably one of their team members put it up without asking the team and then the rest of the team asked that member to remove it.
jager1976
07-01-2009, 14:55
There is another rule stating that the wheel's rotation MUST BE inline with movement of the bot, so that the axle the wheel is on is perpendicular to the bot and wheels movement
I'd be interested in where this rule is. <R06> says "The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub). The ground being a plane, this seems to suggest that the wheels may be angled. Maybe?
FatBabyJezus
08-01-2009, 01:25
would car steering be a bad idea?
repairmanray
08-01-2009, 01:48
if you ask me, we dont have to have the wheels spinning in a circular orbit. by building a hub that allows the rotation of the wheel to be offset from the center, a different driving stlye would result. im not sure if this would help at all, or be legal for that matter, it just seemed interesting.
Kind of like bouncing on the bumper of a truck stuck in mud? Interesting idea.
WildStangs removed the vid. because they are not willing to share their results of the test.
CraigHickman
08-01-2009, 03:08
WildStangs removed the vid. because they are not willing to share their results of the test.
And this knowledge is proven/justified how...? Baseless assumptions aren't good. I'm sure there was a good reason for them to remove it. Whatever it is, it's pointless to speculate until we're told.
That being said: Teams that think in the boxes that we've gotten used to over the last few years WILL fail. I'm willing to bet on it. A standard skid-steer robot will be weighed down and messed up by the trailer. Only robots with specifically designed steering and control mechanisms will be able to move around with enough speed and control to make themselves useful on a strong, competitive level. Just ask anyone who's driven up here in the north about how to drive on Ice, and you might learn some good tips on how to control your bot.
Dracco1993
12-01-2009, 21:04
I'd be interested in where this rule is. <R06> says "The ROVER WHEELS must be used in a “normal” orientation (i.e. with the tread of the wheel in contact with the ground, with the axis of rotation parallel to the ground and penetrating the wheel hub). The ground being a plane, this seems to suggest that the wheels may be angled. Maybe?
The wheels can not be angled as you are talking about.
Dracco1993
12-01-2009, 21:07
As we have found out in tests, it is better to go to full power right away thatn try to slowly speed things up, since kenitic and static friction are so close to each other that it makes no diffrence but it is easier for the drivers to control.
Andrew Bates
12-01-2009, 22:21
As we have found out in tests, it is better to go to full power right away thatn try to slowly speed things up, since kenitic and static friction are so close to each other that it makes no diffrence but it is easier for the drivers to control.
Hmm, your tests seem to disagree with ours...
GaryVoshol
13-01-2009, 07:54
The axle going through the wheel can point north, south, east or west, or any spot in between the cardinal directions. The axle cannot point at an up or down angle.
Einstein
13-01-2009, 15:54
WHEELS WHEELS AND MORE WHEELS. Not in line wheels but hub to hub to hub wheels. like a 16 to 20 wheel drive. all wheels drive all wheels have friction. You need as much surface on the ground as you can to get more friction. ie you have a car with the right inflation on all the tires you get the best gas per mile.
they always tell you to check you tire pressure. now let out 10 lbs what happens to you gas mileage (it goes down right) you did not change the weight of the car you just changed how much surface hit the ground which in turned gave you more friction. and that is what you need more surface to hit the ground. the weigh divided over the wheels don't matter the bot still stays the same weight
sdcantrell56
13-01-2009, 17:40
WHEELS WHEELS AND MORE WHEELS. Not in line wheels but hub to hub to hub wheels. like a 16 to 20 wheel drive. all wheels drive all wheels have friction. You need as much surface on the ground as you can to get more friction. ie you have a car with the right inflation on all the tires you get the best gas per mile.
they always tell you to check you tire pressure. now let out 10 lbs what happens to you gas mileage (it goes down right) you did not change the weight of the car you just changed how much surface hit the ground which in turned gave you more friction. and that is what you need more surface to hit the ground. the weigh divided over the wheels don't matter the bot still stays the same weight
Simply put, this is incorrect. The friction force is equal to the normal force times the coefficient of friction. This has already been discussed numerous times. Increasing surface area on an ideal surface will never increase the friction.
We found out a few interesting things today. we mounted 4 of the rover wheels on the back of our 2008 lowered center 6wd and drove it on the FRP. the weight is biased pretty far back so the front never touched down. right away you could push it around really easy like they did at the kickoff. after wearing out a few batteries driving around, the wheels got a a significant amount more traction. we then put a small scale in front of the front bumper and told the driver to push as hard as he could without spinning. it pushed with about 35 lbs of force until the wheels broke loose, when the scale went down to about 19 lbs.
This type of drivetrain (long 6wd, lowered center) drives like a smaller version of a wide robot as long as it stays on the rear 4 wheels turning was pretty easy but we haven't hooked up a trailer yet.
our team had an awesome idea!!!!!! we are gunna use a high speed fan to create traction because the fan will wanna stick to the regolith when air is being sucked in haha:D
1 more thing i forgot have an awesome 2009 and good luck:yikes:
sdcantrell56
14-01-2009, 06:34
our team had an awesome idea!!!!!! we are gunna use a high speed fan to create traction because the fan will wanna stick to the regolith when air is being sucked in haha:D
This has been deemed illegal by the GDC. Any method of increasing traction is illegal this year, which would include attempts at increasing your normal force.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.