Re: Our Team Needs Help on the Drive!
 

Well throwing another question out there that I have, for crab drive, I have never really seen one so I am not sure how they work or how to set it up. From my understanding this is what it would look like (correct me if I'm wrong)
|\--------/|
|--|----|--|
|=|----|=|
|--|----|--|
|---\__/---|
|-----------|
|oo------oo|
_________

(This is not to scale, (-) is just a space so ignore them)
The circles (o) would be the wheels powered, and pointed to the front of the bot, the (=) would be wheels facing <- at a 90 degree angle, and powered. Is my interpretation of crab drive front wheels correct?

Re: Our Team Needs Help on the Drive!
 

[quote=dtengineering;794539]And we're thinking of something like:

With the center wheels dropped a bit, but all wheels driven. Depending on where the centre of gravity is located it begins to approximate a two wheel drive as the majority of the mass will act through the centre axles. As one approaches this state the transverse friction on the wheels preventing skid steering should begin to approach zero.

Of course, building in the wide direction does present other challenges, but also presents other opportunities. We've never built a "wide" robot before, but this year's game has inspried us to try something new.

Jason
QUOTE]

This is the exact design that we are thinking of. We used this design on our 2006 'bot, and it worked great. Though the wheel layout was different in '06, I think that the new wheel layout will prove worthy. Now to see how it "slides" on the new floor...

Re: Our Team Needs Help on the Drive!
 

It's interesting to see some convergence. These are the drive configurations for our two leading design concepts.

Re: Our Team Needs Help on the Drive!
 

Not quite, but close
If the front two wheels are in a crab modules then they are free to rotate as directed by separate motors. i.e. when turning the crab wheels (=) would be exactly as depicted above, but when moving in a straight line the wheels would be parallel to the rear wheels. The wheels can be turned at a whim using two additional motors. Each crab module consists of a drive motor (generally a CIM) and a turning motor which turns the module.
Here are some threads and images better explaining crab drive:
http://www.chiefdelphi.com/forums/sh...ad.php?t=55454
http://www.chiefdelphi.com/media/papers/1836 (he calls crab swerve, but they are very similar)

Crab drives are a challenge to build, but can dominate if correctly implemented.

Re: Our Team Needs Help on the Drive!
 

Okay our team officially decided on the following:
______--------------______
|--------\------------/-------|
|---0-----\_______/----0---|
|-----------------------------| ******28*******
|-----------------------------|
|--00--------------------00--|
|-----------------------------|
_______________________

********38*********


So the (0) is a wheel pointing forward only, and it is a tank drive. The width is on the side, and the longer side, in the back. All wheels are powered by one chain on either side, one motor on either side.


This was what I tried to get our team to do:

______--------------______
|--------\------------/-------|
|---=-----\_______/----=--|
|-----------------------------| ******28*******
|-----------------------------|
|--0----------------------0---|
|-----------------------------|
_______________________

********38*********

The (=) is a crab drive wheel and the back two wheels are just pointed vertically, all wheels powered. Each wheel would have it's own motor and chain moving it.

So if the first drawing is A, and the second one is B, then here are some questions:
1) Which drive looks better for this competition?
2) What are the pros and cons of A?
3) What are the pros and cons of B?
4) Suggestions that we should take under advisement, and any other comments.

TY to all who respond!:]

Re: Our Team Needs Help on the Drive!
 

The crab module will be better for the competition provided it is constructed properly and the programming is done correctly. However, the crab drive may result in disaster if things go wrong.
A is more easily constructed and will provide more time for driver training testing etc. It should perform almost as well as B.
B will be more maneuverable and better able to push in all directions; however it will be much harder to build, design, program.
If you have never constructed a crab drive I would recommend you build a tank drive as in A and try to design a crab module in the off season. Too much can go wrong during the build if you attempt to try something completely new in a single build season. Try to make what ever you decide the best it can possibly be. Oh and have fun.

Re: Our Team Needs Help on the Drive!
 

1. It depends on the location of the cg and whether the wheels are coplaner or some set is dropped. The trailer is effectively part of the robot.
2. See (1). What do you expect the doubled wheels to do for you?
3. See (1). What do you expect the crab wheels to do for you?

4. I believe that what the GDC is trying to do with this years game is to force teams to do the physics and math and understand all of the forces at work. They have effectively created an environment for which past rules of thumb and past experiences and intuition may not apply. But the laws of physics still apply. Use them to engineer your robot. As you get to understand the physics insights will start to appear.

OK, you have a couple of potentially viable configurations. Put the trailer in the picture. Stick the cg in some realistically attainable envelope and start calculating with the cg in the center of that envelope. What net force will the robot have? What turning moments will the robot have? Move the cg to the corners of that envelope and recalculate. Does it still work?

The traditional steps in evaluating design concepts are:
1) Does it work well on paper. If not, it still might work in the real world but it's usually not worth the risk. Engineers don't build things that don't work on paper (unless the're software engineers ;) )
2) If it works well on paper, confirm that your assumptions and calculations are valid and that it is likely to work in the real world as well.
3) Is it feasible to complete in the allotted time, i.e. detail design, build, test, etc.
4) Does it meet all the requirements of your application (i.e.) your selected strategy?
5) Is it likely to be competitive? (in the commercial world this is often a life or death parameter. In government nobody cares. In FIRST everybody is a winner.)

It's still early. Do the math. If you need help with the physics or math just ask.