Wheel Placement on Robot

Hello,

I'm building a 4WD robot, but I'm a little confused about how to optimize the robot for turning. I need to turn the robot 90 degrees, and I need the center of the robot to stay in the same spot as it turns. Anyway, were should I place the wheels? Should I put them far from each other, or as close a possible? The Robot will be 2ft wide and 3.5ft long.

Thanks!

are you using a servo or motor on each wheel?

One 12v, 3 amp (no load) motor with 6in wheels.

in my opinion i would say have the wheels near the corners with the transmission in the dead center of the robot if you want to keep the turning radius in the center

but in that case you might just need a motor for each side of the robot

Sorry I wrote that wrong...

One 12v, 3 amp (no load) motor with 6in wheels.

One Motor for each 6in wheel.

than possibly not sure if you have the access to scrap wood or not but you might want to make a couple of wood platforms with those dimensions and test different mounting positions for the wheels that way when you found a place that works with the radius you want you will have the measurements and everything (yes i learn a lot by trial and error)

Bots like below operate like skid steer equipment, which operate similar to tanks. The wheels on the same side operate together.

Hope those motors are fairly powerful. Hope it won't be run on carpet (unless you don't care about your carpet long-term).

What this is called is "skid-steering" (actually differential steering) - meaning the wheels skid sideways as the vehicle turns:

In fact, your dimensions seem really close (scaled) to that of many skid-steer loaders like depicted in that link above. Regardless, it really tears up the ground, and if used on hard surfaces (like concrete and such) that have no "give" - it can mess up bearings, transmission components, and tires in short order. Keep that in mind for your robot. If you plan on this being used on dirt or grassy surfaces only, all the better. But if being used on hard surfaces (especially with grippy tires and lots of weight), expect the components to fail quicker than normal.

The best orientation (as far as wheels are concerned - tracks are a whole 'nother matter) for the wheels are at the extreme corners of a square base (that is, at the points where, if a circle was circumscribed around the square, the vertices of the square would meet the line of the circle). Ideally, you would also be able to angle the wheels such that they tracked the circle as the vehicle turned in-place. This of course would be more complex to implement.

Thanks for the replies.

Ideally, you would also be able to angle the wheels such that they tracked the circle as the vehicle turned in-place

If your talking about what is in the image below, it would be to hard and cost to much to add that into the robot I'm working on.

than possibly not sure if you have the access to scrap wood or not but you might want to make a couple of wood platforms with those dimensions and test different mounting positions for the wheels that way when you found a place that works with the radius you want you will have the measurements and everything

I will give that a try.

The best orientation… wheels are at the extreme corners of a square base

I will try that out on the wooden prototype.

Regardless, it really tears up the ground, and if used on hard surfaces

It will be running on hard surfaces (the road) but I'm not to worried about messing it up as it looks strong...

http://www.parallax.com/product/27971

Thanks!

400px-Pivot_module_section_2012.jpg

My robot is very similar in size, has similar sized wheels and runs on grass.
My experience:
I'm still looking for a motor that lasts longer than a couple of weeks :frowning: I have killed lots of motors of several types in the last years :frowning:
I have upgraded the rods to the wheels from 6mm steel to 6mm rust free steel to 10mm rust free steel; because they broke ( sometimes in a couple of hours) due to the turning.
I ended up using 2 ball bearings in a aluminium case (the plastic ones broke) to support each wheel rod.
The motor is tightly attached to the wheel rod but it is flexibly attached to the alu frame. This means that the motor can not turn around its axe (and as such can deliver power) but it can move when the wheel is forced up/down/left/right (so it does not take any axial or radial force except for its own weight induced force).
So from experience I'm fully with Cr0sh here.

I'd be interested to see the wheels you are planning to use as I'm not fully happy with the wheels I'm currently use and I hardly find 6 inch wheels.

Best regards
Jantje

I posted the link to them above, but I will post again.

http://www.parallax.com/product/27971

Och I wasn't aware it was a motor wheel assembly.
This looks good on paper but the device seems to be under revision which sounds bad to me.
Best regards
Jantje

If turning is difficult, the bot might be hinged in the middle and rotated left/right to help with turning. Another setup below.

This looks good on paper but the device seems to be under revision which sounds bad to me.

I have been in contact with the company and they have had the kit for years, however, their supplier for the motors just stopped making that model. They found comparable motors, but the mounting is slightly different so they are working on a new mounting block.

If turning is difficult, the bot might be hinged in the middle and rotated left/right to help with turning. Another setup below.

I think making the robot swivel like that would just be to hard and time consuming for this bot, but that could be fun to do in the future with a smaller robot!

A lazy=susan like below could be used in the center of the bot to allow the flexing of the front and rear sections for turning. For skid steering keep the wheels close together for easier pivoting and turning.

http://www.homedepot.com/p/Shepherd-6-in-Lazy-Susan-Turntable-9548/100180572#.UiDGFynD_ix

The best orientation… wheels are at the extreme corners of a square base

and

keep the wheels close together for easier pivoting and turning

Now I'm confused... keep the wheels as close as possible or as close to the corners as possible? I have a feeling there is only one right answer for this. Personally I would think as close together would be best, but I really don't know.

Also, were should I put most of the load (12lb of batteries) over the wheels?

As far as having the robot pivot it would just be to much for this robot. I know it is possible, but time wise I don't have time to redesign the frame of the robot which is probably already assembled. (i'm having somebody weld the frame together). However, that Lazy Susan Idea looks cool so I will add that to my "when I get some extra time and money" folder.

weight wise best placement of the load is in the center

Now I'm confused... keep the wheels as close as possible or as close to the corners as possible? I have a feeling there is only one right answer for this. Personally I would think as close together would be best, but I really don't know.

A lot may depend on the size of the wheels compared to the size of the platform. Might be worthwhile to start by attaching the motors/wheels to a square piece of plywood and test the turning performance. Larger bots might benefit with some type of suspension if the bot will travel over uneven ground.

Think about it a bit: We are talking 4 wheels here. If you moved the wheels in the front back, and the ones in back forward, until they were near the center (but still spaced opposite each other), then yes, it would be easier to turn. Heck - if the wheels "merged", then you have the classic differential steered robot (add some casters on the front and back to balance).

Regardless, though, when the robot pivots about its center, those wheels will describe a circle. The closer the wheels are to the line that runs thru the center of the circle, the less drag/skid they will exhibit (and in the case of the classic design, little drag at all - but there will always be some drag, as any wheel that is following a curve, the inner edge /must/ move slower than the outer - and so you have wear; which is why when you car doesn't have proper alignment, your tires are messed up in all sorts of different ways - which is why, even with perfect alignment, you must rotate your tires, to even the wear).

Therefore - if the wheels are going to describe a circle anyhow - with a 4-wheel arrangement, the configuration with the least amount of drag (but more drag than the classic differential two-wheel design, of course) will be where those wheels contact the circle they are describing - which has to be a square (and, again, ideally the wheels could pivot - but as you noted, this would drive up complexity and cost, which I also realized when I posted that originally; I was merely suggesting the ideal way to minimize drag on such a 4-wheel arrangement, irrespective of complexity or cost).

Now - you may wonder why the various skid-steer equipment makers don't make their machines square? Well, it mainly has to do with the purpose of those machines and where they are operated. They are meant to be operated on surfaces with "give" (indeed, the manuals for these machines generally explicitly say not to operate them on concrete, asphalt, or other hard surfaces), so that they can easily slide without taxing the drive and other mechanical components. They are also generally designed to be small, while still carrying a good sized load, and be maneuverable, which limits making them square. Engine layout and other design needs probably also factor into the reason why they aren't square.