Crab Steering on Regolith
 

In Lunacy this year, will a crab steer be allowed in the form of a brake?
I know the rules specify that all wheels must be in orientaion with the robot, but with crab steer, you could possibly turn your wheels to stop you from gliding more efficiently before you change directions.

Is this a legit idea or do the rules not permit it?

Re: Crab Steering on Regolith
 

<R06> Seems to create a problem with any type of steered wheel this year: if the axis of rotation of the wheel has to be parallel to the ground, then you can't rotate the wheel about a vertical diameter, as you typically would while steering. However, according to the rule you could rotate the wheel about a horizontal diameter and make use of the higher transverse coefficient of friction do propel the bot. I hope we see some clarification of <R06>.

Re: Crab Steering on Regolith
 

The only rule I'm aware of which could limit it is R06, and as I read it I don't see any problems with your idea.

However, you won't get more stopping force from that since the friction with the ground would still be equal. My guess would be that you'd have better luck either running the wheels in reverse, or at least equal results by using encoders and some simple programming to prevent the wheels from moving when you want to stop.

Re: Crab Steering on Regolith
 

horizontal axis of rotation just means that you have to put the wheels flat on the ground. the horizontal axis is your axle so you can spin your axle on a horizontal plane .

Re: Crab Steering on Regolith
 

As i interpret this rule the key part is

The axis i believe they're referring to is the axle. So then I interpret it as saying the wheel must remain vertical, relative to the ground, but doesn't necessarily need to maintain its orientation relative to the robot chassis.

I could be misinterpreting this, so if it a serious issue for your design it would be best to contact FIRST for clarification.

Re: Crab Steering on Regolith
 

why not try using a massive flywheel and rev it up to a couple thousand rpm, next turn the flywheel with your windsheild wiper motor in the direction you want to go, see how the robot reacts to this inertia, if it does not turn on a dime in zero friction enviornment why would nasa use it on their shuttles and other vehicles?

Re: Crab Steering on Regolith
 

This is known as a "reaction wheel". You should research if it will provide more turning capability than what you can already do with a standard wheeled drivetrain. If it will, then use it, but if it won't....

EDIT: Oh, typically you would use a reaction wheel horizontally, speeding it or slowing it rapidly to transfer the angular momentum to the robot chassis. What you propose is a simple gyroscope, which could also be effective.

Re: Crab Steering on Regolith
 

I was wrong... about the flywheel, still think it's risky


Secondly, the crap drive concept seams legal, since the wheel is turning about a horizontal that would be parallel to the playing surface, the wheel is able to turn becuase it whole drive assembly/input power shaft can turn with it. Those rules make it so teams don't lay wheels on their side (axis of rotation is perpendicular with the plane of play), or thats what I think.

Re: Crab Steering on Regolith
 

I am not using it to acelerate but to turn in tight conditions, the primative skid (coded) tank steering or omni steering is old news and the wheels would skid out of control. this steering works in SOLIDWORKS and it can cause the robot to turn about an axis without interacting with the ground but under no circumstances will it acellerate the robot.

Re: Crab Steering on Regolith
 

It hasn't seemed to have been mentioned here, so I'll point it out.

You may wish to consider that especially on this playing surface, the orientation of the wheel doesn't really matter when it comes to putting on the breaks against a pushing robot. Any of the motors generally chosen for the drive train of a robot are more than capable of holding the wheels locked (not rotating) when pushed on this surface - the friction between the wheel and playing field is much less than the force required to overcome the stall torque of the motor. That said, the orientation of the wheel to the direction of undesired motion doesn't really matter.

Additionally once you are sliding, it will be very difficult to regain control of the robot by slamming on the gas in the direction you want to go. Your already sliding wheels would still have very little traction, so it would take a while to regain control of your robot.

Re: Crab Steering on Regolith
 

So this hasn't really been debated yet but would a crab drive even be particularly effective on this surface?

It seems to me that the lack of friction would make it so that turning the wheels would not really result in a change of direction quickly. It could actually be worse since the wheels would be skidding and thus kinetic friction would apply.

Re: Crab Steering on Regolith
 

Note that the published values for the wheel's transverse coefficient of friction are almost twice that for the inline coefficients.

In other words, yeah... turning the wheel 90 degrees to your direction of travel will bring you to a stop more quickly than just stopping the wheel.

The downside is that you will be in a skid at that point and will have sacrificed directional control.

Jason

Re: Crab Steering on Regolith
 

So based off of the transverse frictional value a crab drive could actually be effective. It seems like the programming might be even more difficult than a normal crab drive though to get full performance from the system.

Re: Crab Steering on Regolith
 

How would the direction of the wheels dictate the friction any more than normal? Technically your "contact patch " doesn't grow or shrink any, and the wheels have no lip (read: they're rounded), so you don't have any edge to dig in to.

Re: Crab Steering on Regolith
 

That is something I dont understand either. The wheels are molded from a uniform material and there is no tread pattern so I dont see how friction would be different. Maybe it is due to the plastics grain or something like that.

Re: Crab Steering on Regolith
 

you need to use momentum sustained in an independent body from the robot and then transfer the momentum to the robot from the independent body.

Re: Crab Steering on Regolith
 

You need to hit a wall or another robot? :D

Re: Crab Steering on Regolith
 

Yeah, I was playing with a wheel yesterday and the only non uniform part i noticed was the grain of the plastic, to me it seemed like there was a substantial difference between the wheel sliding inline with itself and it sliding along its axis


Well, now that I think about it, would just running your wheels backwards give u better stopping than turning your wheels sideways???

Re: Crab Steering on Regolith
 

I haven't got my hands on a wheel yet, but those who have tell me that it is slightly rounded to give a "high centre".

This could cause the change in friction published in the manual section 10.2.4.1

Jason

Running the wheel backwards (dynamic inline friction... 0.5) would be worse than just turning the wheel sideways (dynamic transverse friction 0.10) according to the published values.

Re: Crab Steering on Regolith
 

Do you mean .05? or 0.5? for the dynamic inline friction?

I'm no physics master, but I'll take a guess. :yikes:

Could this have anything to do with the coefficient of static friction being greater than that of kinetic? That is, when the wheel is in the forward direction it is rolling and therefore applies static friction, while the wheel is turned perpendicular to the direction of travel it slips and applies kinetic friction. Feel free to correct me if I'm wrong.

I don't know if this theory works because the wheel may also be slipping while the robot goes forward as either way there is barely any friction.

Re: Crab Steering on Regolith
 

i believe that crab drive would be allowed because rule R06 reffers to the "normal" orriantation as the wheels being flat on the ground. this only means that the wheels cant be angled to the floor, it has no constraints on how the wheels can be in oriantiation to the robot.

in short, rule R06 states that the wheels have to be perpendiculer with the floor, therefore crab drive would be possible.

Re: Crab Steering on Regolith
 

Just a little advice for everyone....

Saying "I Think" doesnt mean anything. Right now the only form to find what is allowed and not allowed is the manual. Once the Q&A and the update system is started you should get things cleared up. If you get to competition and the inspectors say something is not allowed and you say you thought it was ok they arent going to care. If you think of ANY possible controversy over a strategy you want to use, look it up in the manual and be positive it is in compliance with all the rules.

Also, rules from past years games, or other sports dont apply. Dont assume things are the same as last year.

Joey

Re: Crab Steering on Regolith
 

<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)...

The floor is a plane. The axis is a line. A plane is parallel to a line if they will never touch if their length was extended to infinity. This means the axis which the wheel is driven about must always be in a horizontal orientation. The purpose of this is to disallow tilting of the wheel. Therefore, R06 does not identify crab steering as an illegal system.

On another note, I also thought that crab steering would work, but for a different reason. In a car, drivers are told to turn with a skid when driving in slippery conditions. This way, the tire can grab and begin rotating with the direction of travel. I do not think that any type of steering will prevent sliding but I think rotating the wheels in the direction of the slide will allow the robot to regain traction faster. But this paragraph was just my thoughts, nothing which is absolute fact. As was mentioned earlier, no statement starting with "I think" should be considered as true without further investigation.

Re: Crab Steering on Regolith
 

That's all fine and dandy except for the fact that different people interpret rules differently. Finding out other peoples' interpretations can help people to make more informed design decisions before the Q&A starts. Reading the rule book can only get you so far.

And on a note pertaining to the actual thread....

I think (haha) that a crab drive has its advantages and disadvantages in any situation. In this years game, however, I think that the difficulty in controlling is already going to be such that it discourages a crab or swerve drive. Granted, we haven't leaned towards crab drive in the past, but this year while brainstorming we didn't even consider it. I'm sure some team out there will pull it off, but it'll take a pretty skilled drive team. I do encourage you to try however; that team might be yours. ;)

Re: Crab Steering on Regolith
 

Thats exactly my point. Interpruitations dont matter. The only thing that matters is what is stated in the rule book. I am bringing this up because I was a ref at our state lego tournament. We had a problem with the INTERPRUITATION of a rule that was allowed at various regionals. We looked in the manual, and there was a clear rule stating what was previously allowed is not allowed. I'm just saying dont guess on this type of stuff. Take the initiative to look rules up your selves, and if a rule is not clear, submit a Q&A, or keep up with the updates. If you get questioned on a rule at a competition and you say "well I heard on CD......." thats not going to cut it.

I was only talking about saying I think when questioning the legality of something. Offering opinions about strategies is fine.

Joey

Re: Crab Steering on Regolith
 

But before the Q&A starts during the first two critical days of strategy and design, a team needs to have guidance on wishy-washy rules. Knowing others' interpretations of those rules helps them get a general idea of the general consensus of the rule. I'm not saying they should stick with this interpretation of the rule throughout build season (they should certainly confirm it with updates and such), but an unofficial interpretation of a rule that prevents a standpoint in design is better than an official rule that causes many conditions to be built into designs and strategies.

In any case, this isn't the thread for this discussion. I apologize for taking up space. :(

Re: Crab Steering on Regolith
 

You are forgiven

Re: Crab Steering on Regolith
 

My team was actually looking to use a crab/swerve drive for the first time this year. Upon seeing that we would be pulling a trailer, I began wondering how that would affect changing direction. Going forward was obvious, you are just pulling. But if you go left or right to avoid a robot or get where ou want to be, would the trailer cause the robot to do kind of an awkward rotation while it moved? And the trailer would jack-knife when trying to go backwards, would that cause the robot to want to follow it somewhat?

Re: Crab Steering on Regolith
 

I'm also very interested in some sort of crab drive for several reasons.

First of all, using a style of steering such as a tank drive, requires skidding to turn, automatically meaning you lose something like 20% of your traction. By utilizing a crab-drive style steering it could allow you to turn without skidding more easily.

It would also allow for a lot of flexibility, since just because you have the capability of turning each wheel, it doesn't mean you have to. I can see where multiple drive setups could be tested by modifying the way the controls operate. Three main ideas that have crossed my mind so far are a conventional car steering, a simulated tank-drive, and also a monster-truck type steering. All things that could be played around with.

I think that flexibility could be a very useful this year, as it may aid drivers to a great extent in being able to attain good control.

Re: Crab Steering on Regolith
 

I don't know, so here's my $0.02.

In years past, and I can already see it this year, people read too much into, and "lawyer" the rules. The rules are all there for a reason, and for the most part, the intent of each rule should be more or less obvious. The Q&A typically responds in line with the INTENT of the rule, and thusly, thats the basis we should be going on.

To me, the INTENT of <R06> is to prevent teams from utilizing the spokes of the wheel as a traction surface. It is NOT (IMO) intended to stop teams from creatively USING these wheels, in, for instance, a big omniwheel made with these as the rollers, or other creative uses. My interpretation is that the "normal" way to use a wheel is with the tread in contact with the floor. However, I'm by no means on the GDC, and thats for the Q&A to make the final call.

The fact that an omniwheel with these as the rollers would be enormous is somewhat irrelevant to the point.

Re: Crab Steering on Regolith
 

I totally agree. they just want you to use the "flat" part of the wheel.

Re: Crab Steering on Regolith
 

A big omniwheel would be redundant as they are already basically omniwheels. A mecanum (spelling?) wheel however...

Also I'm not sure a giant omni wheel made of these would work because then technically your wheels don't always have their axles parallel to the ground, but they do when they are in contact with the ground...Still questionable.

Re: Crab Steering on Regolith
 

Actually, these wheels are anti-onmi wheels. The idea behind omnis is that they have a negligible transverse coefficient of friction (CoF) (due to the rollers) and a much higher inline CoF (usually due to the rubber material of the rollers). These wheels are the exact opposite, with an inline CoF on less than half of the transverse CoF. In fact, I doubt a standard configuration of omni wheels would move at all, with the exception of turning in place.

Re: Crab Steering on Regolith
 

I am not sure if this has been said but I believe that somewhere in the rules it states that the coefficient of friction doubles if the wheels are turned sideways