4WD Turning Difficulties

[Brandon Holley]

Originally Posted by Joe Ross

We had far more trouble turning from the lack of traction then from too much transverse friction.

http://www.youtube.com/watch?v=u8aqjM1mmEs

thanks for posting!

[Gabe Salas Jr.]

Originally Posted by Joe Ross

We had far more trouble turning from the lack of traction then from too much transverse friction.

http://www.youtube.com/watch?v=u8aqjM1mmEs

The surface we were running on has about double the coefficient of friction as the real surface.

I'm more interested in seeing it turn with a trailer.

[GregW11]

Hm... what if you had a six-wheel design (skid steer, it won't work so well for arcade style) in which the front and back wheels had motors but the center wheels didn't, and there were sensors in all the wheels. The center wheels would be the actual rotation of the wheels while the front and back would be what the motors are doing, and the controller would adjust the output of these motors based on the actual wheel rotation versus what the motors are trying to do. (if that made any sense at all) I'll try to explain in a diagram below:

| | <motors on these wheels sensor compares rpm to center wheels

| | <no motors on these wheels, sensor gets rpm for actual speed

| | <motors on these wheels, sensor compares rpm to center wheels

Controller will adjust actual output of the motors. It's not fully autonomous, since you're still controlling which direction it is supposed to be going and the speed desired.

If anyone wants me to I'll try and do a sort of block-diagram showing a general flow of possible commands.

[Craig]

much appreciation for the vid. love to see a trailer contributing some drag and then see how little steering happens.

lets see who can put one together and post her up

[Mike Betts]

Joe,

Thanks for posting the video... The problem you had turning at about 15 seconds into the video reinforces a worry I have had... From what I could see, I believe it was caused by having a different COF for the right two wheels as compared to the left two wheels.

If you are using a 4WD (skid steer) and you get two wheels on the carpet at the field perimeter, you may never get back onto the main playing field...

Mike

[Joe Ross]

Originally Posted by Mike Betts

Joe,

Thanks for posting the video... The problem you had turning at about 15 seconds into the video reinforces a worry I have had... From what I could see, I believe it was caused by having a different COF for the right two wheels as compared to the left two wheels.

If you are using a 4WD (skid steer) and you get two wheels on the carpet at the field perimeter, you may never get back onto the main playing field...

Mike

We were worried about that too. I think in the video was the first time we got into that situation, I don't remember having that much trouble when we did it again. We still do want to do a lot more tests. I think this year more then ever, teams will want to get a base driving as soon as possible. Trying to drive for the first time with 5 other robots on the field will lead to trouble.

[Joe Johnson]

Originally Posted by GregW11

Hm... what if you had a six-wheel design (skid steer, it won't work so well for arcade style) in which the front and back wheels had motors but the center wheels didn't, and there were sensors in all the wheels. The center wheels would be the actual rotation of the wheels while the front and back would be what the motors are doing, and the controller would adjust the output of these motors based on the actual wheel rotation versus what the motors are trying to do. (if that made any sense at all) I'll try to explain in a diagram below:

| | <motors on these wheels sensor compares rpm to center wheels

| | <no motors on these wheels, sensor gets rpm for actual speed

| | <motors on these wheels, sensor compares rpm to center wheels

Controller will adjust actual output of the motors. It's not fully autonomous, since you're still controlling which direction it is supposed to be going and the speed desired.

If anyone wants me to I'll try and do a sort of block-diagram showing a general flow of possible commands.

Unpowered wheels with significant weight on them means that the powered wheels have just that much less normal force to get traction.

I believe that successful teams are going to optimize their traction. Don't be fooled by past years.

In the past you saw robots jumping as they turned corners, drive motors that cut out from the circuit breakers tripping and sometimes smoke pouring out of the motors and you knew, ah, something is less than optimal there. This year there will be none of that. Teams will think, my robot is not hopping as it turns, my motors are not tripping the breakers or even getting hot... ...everything is right with the world.

No such luck.

Many teams will not discover their lack or RELATIVE mobility until practice day. Too late.

By the way, backing up long distances is going to be impossible for all but the best teams. Think of the mobility issues that is going to cause.

This is a year I regret more than ever being on the sidelines... ...the engineering challenges are fun, not well understood and provide significant advantages...

...now if only I weren't worried about loosing the audience...

Joe J.

What about using 8 wheels, 4 on each side. And there would be one motor for a set of two wheels, or one motor for each side. Would that work better or worse that just a straight 4 wheel drive train?

[jimbot]

What about a 4wd car steering system? the only problem i can see is that you could encounter difficulties in backing up if you needed to.

[T3H_K3YM45T3R]

what about a drive train like that, have it set up with duals in the back and put them close together, technically still 4wd, but set up like a semi almost. this is what my team is going to try.


| [ . . . . . . . . . ] |
| . . . . . . . . . . . .|
| . . . . . . . . . . . .|
| . . . . . . . . . . . .|
| . . . . . . . . . . . .|
| . . . . . . . . . . . .|
| . . . . . . . . . . . .|
| [[ . . . . . . . . ]] |
| [[. . . . . . . . .]] |

[ZakuAce]

I was thinking about all this discussion of trailers. Would it be possible to turn the robot without involving the trailer? Like have the hitch on some sort of rotating...I'm not sure what the right word is, "cuff" comes to mind I don't know why...so the robot turns but the trailer stays stationary? Then the robot moves foreword and the trailer is pulled into position behind the robot.

Edit: Also I completely agree with the car steering drive train for a 4wd. It has the advantage of keeping your drive wheels in line with the direction you want to go. Even if you skid, your drive wheels should be able to pull you back on track, like a car on ice. A skid drive means all the wheels face the same direction, giving you less control.

Edit 2: I wouldn't worry so much about having to back up, this is a small problem compared to getting your entire robot to maneuver correctly.

[SWIM]

However, according to
http://www.chiefdelphi.com/forums/sh...07&postcount=7
the coefficents of friction in real life happen to be somewhat different from what FIRST tells us they are.

If you run the spreadsheet with the coefficents of friction being .12 in every direction, there is a good chance that your robot will be able to turn

Quite easily, in fact. Although, you should keep in mind that "turn" and "rotate" aren't entirely synonymous. If you attempt to turn with any sort of speed, you're probably going to instead rotate in place while continuing to slide sideways, so drivers of skid steer 'bots are going to have to practice their four wheel drifts to be able to maneuver effectively this year.

[menace101]

I don't remember where, and it's getting late for me, so I don't want to look it up, but I think, THINK, that there is a rule stating that the place the trailer attaches to has to be stationary or something.

And I would like to bring up the 4WD drive with the steering similar to a car, what's that called, by the way? I think it would definitely give us more control overall. Especially when compared to skid steering or something. One question that bothers me about it though, will the low coefficients of friction render this kind of steering just not worth the time input into designing and making the system?

[SWIM]

Originally Posted by menace101

And I would like to bring up the 4WD drive with the steering similar to a car, what's that called, by the way? I think it would definitely give us more control overall. Especially when compared to skid steering or something. One question that bothers me about it though, will the low coefficients of friction render this kind of steering just not worth the time input into designing and making the system?

I believe the FIRST colloquial term would be "swerve drive". edit: apparently not

The major obstacle in your way is probably going to be replacing the chain-drive with shafts and gears. Also, CV joints would likely be involved. Making the steering rack it's self wouldn't be too bad, but the whole system would add a couple layers of complexity to what needs to be the most reliable part of your robot.

While turning, you'd be able to make more efficient use of your traction, and it would be far more controllable at speed. Whether that's worth the extra complexity, weight, and motors depends on what your team finds important in a drivetrain.

edit: come to think of it, you could use a single CIM to drive all four wheels, and a CIM to power the steering rack. that could potentially be a pretty good solution, but you'd need two or three limited-slip differentials for it to handle properly, and I think andymark is fresh out of those...

[Racer26]

Originally Posted by menace101

And I would like to bring up the 4WD drive with the steering similar to a car, what's that called, by the way? I think it would definitely give us more control overall. Especially when compared to skid steering or something. One question that bothers me about it though, will the low coefficients of friction render this kind of steering just not worth the time input into designing and making the system?

"Car" style steering is known as Ackermann Steering (front wheels steer), and the variation (like a monster truck, and some luxury cars have) is called 4WS (4 wheel steer)

"Swerve", or "Crab" drive systems are where the wheels are rotated to point in the direction you want to go, (typically they can be turned at least 180deg so you can get the full gamut of translational possibilities).

"Kiwi" drive is using 3 omniwheels spaced 120deg apart traditionally, but in this years game, it would likely be possible with the rover wheels, you drive the front two toward each other to move forward, and steer with the rear one.

"Holonomic" drive is using 4 omniwheels spaced 90deg apart. By varying the speeds, you can achieve translation in any direction. I expect this too would be possible with the rover wheels

"Mecanum" drive is using mecanum wheels (like omniwheels, only the rollers are on 30-45deg angles instead of 90s, and you mount them like a "regular" 4wd setup). These too can be used to achieve translation in any direction. They're not usable this year in any way (well, unless, i suppose if you were to build one using rover wheels as the rollers).

Those, plus 4wd, 6wd, and 6wd-dropcenter are the most common drive systems seen on FIRST robots (well, kiwi isnt exactly popular, but you sometimes see it mentioned). There are others as well, but these form the bulk of whats out there.