Lateral Movement

[MetalJacket325]

Originally Posted by magicode

Because if mecanums do lose force when driving straight (we're still debating), then you're not getting the full force of 4 motors. If you want that and the ability to strafe, an H drive works well.

My team's robot used an H-drive for gateway and it worked great. However, I tink that if you are going to have some sort of large hopper for the sacks, an X-drive or mecanums would be better because they don't require the extra space of the H-drive's strafe wheel(s).

[magicode]

Yeah, I liked X/ + / mecanum drives for the last 2 games better than H drive personally (and they're more fun to program, but maybe I'm just weird). I was just explaining the rational being using H-Drives vs X / + / mecanum drive.

[dontworryaboutit]

ok so I thought about something while mowing the lawn that I said in the other debate thread and I thought it was worth restating.

If the rollers on the wheels rotate:
There are movement components both parallel to the direction of the wheel and parallel to the axis of rotation of the wheel.

The component parallel to the direction of the wheel can be accounted for (by the rotation of the wheel).

The component parallel to the axis of rotation of the wheel cannot be accounted for because the wheels are not moving parallel to the axis of rotation of the wheel (they are restrained in that direction by the robot frame).
Therefore, the rollers can only be spinning if:
1) The robot is moving laterally (it is strafing).
-or-
2) The rollers are slipping. This means that the movement component parallel to the direction of the wheel is also slipping (it can't slip in one direction only; if it is slipping, it is slipping in both directions). The result of the component parallel to the direction of the wheel slipping is that the wheel itself loses traction, which happens when friction does not supply a large enough reaction force to the torque delivered to the wheel by the motor.

[Now a continuation of that train of thought, but focused on why it is 30% easier to apply too much torque to Mecanum wheels and cause them to slip]
The frictional force is directly proportional to the weight of the robot. Because the surface of the wheel where the wheels contact the floor (the bottom-most roller) is free to slip in a direction 45 degrees from the direction of the force applied by the motor onto the rollers, the component of the frictional force in the direction of the force applied by the motor onto the rollers is only 70% the total frictional force, meaning the robot will slip with 70% less force if it is using Mecanum wheels as opposed to wheels (such as omni wheels) that have the same coefficient of friction, but no rollers that have a component of their axis of rotation parallel to the wheel's axis of rotation.

I would point you towards my post back in November to illustrate what I tried to explain here:
http://www.vexforum.com/showpost.php...4&postcount=15

[Vex Mundi]

I was just thinking I should jump into this thread to help dontworryaboutit fight for the forces of light but then I saw this:

Originally Posted by dontworryaboutit

ok so I thought about something while mowing the lawn that I said in the other debate thread and I thought it was worth restating.

If the rollers on the wheels rotate:
There are movement components both parallel to the direction of the wheel and parallel to the axis of rotation of the wheel.

The component parallel to the direction of the wheel can be accounted for (by the rotation of the wheel).

The component parallel to the axis of rotation of the wheel cannot be accounted for because the wheels are not moving parallel to the axis of rotation of the wheel (they are restrained in that direction by the robot frame).
Therefore, the rollers can only be spinning if:
1) The robot is moving laterally (it is strafing).
-or-
2) The rollers are slipping. This means that the movement component parallel to the direction of the wheel is also slipping (it can't slip in one direction only; if it is slipping, it is slipping in both directions). The result of the component parallel to the direction of the wheel slipping is that the wheel itself loses traction, which happens when friction does not supply a large enough reaction force to the torque delivered to the wheel by the motor.

This is exactly what I wanted to say, only explained much better .

We can assume that the wheels do not slip, at least under light driving conditions. This means the rollers cannot spin while the robot is driving forward because if they did then the wheels would move either towards or away from the centre of the robot. They are constrained not to, so the rollers do not spin.

As for what force is required to make mecanum wheels slip, that's something that would have to be determined empirically. I assume it's similar to the force required to slip omnis, since they have similar pulling strength. (dontworryaboutit's calculations that it takes 70% of the force that would be required to make them slip parallel to their roller axles seem to check out).

[LegoMindstormsmaniac]

Thinking about the ways that slipping would show -- would it not be visible when being in a defensive battle with another robot? If so, have you all watched this video http://youtu.be/OCll0hrBmRc? Watch at ~1:15 for a pull test VS an all-omni robot. To me, it suggests that VEX Mecanum Wheels and VEX Omni Wheels are (for the most part) equal in a traction/slippage standard. Perhaps I am wrong in thinking that? Perhaps we need to conduct a test similar to this with some slightly heavier, higher-powered robots?

Or perhaps I have completely misunderstood the reason for this debate and have only made myself look like a fool? (If so, I appologize, and please ignore this. :P)

~Jordan

[magicode]

I don't think either side is going to be happy until someone does an analysis of video as I suggested before, or actually draws a line on a mecanum roller and makes a video of it driving forward.

[therealcedz]

Originally Posted by magicode

I don't think wither side is going to be happy until someone does an analysis of video as I suggested before, or actually draws a line on a mecanum roller and makes a video of it driving forward.

The rollers on the mecanum wheel do not turn when driving forward or backwards. If its easier to understand, imagine a C channel with a bunch of meccanum rollers mounted on at 45 degree angles. If you restrict the movement to only forwards and backwards, you are unable to push it forwards or backwards due to the angle of the rollers. Now if you bend the C channel into a circular shape, then in the same principle, the "wheel's" rollers do not move when restricting the movement forwards and backwards.

[dontworryaboutit]

Originally Posted by magicode

I don't think either side is going to be happy until someone does an analysis of video as I suggested before, or actually draws a line on a mecanum roller and makes a video of it driving forward.

I tried to take a video on Thursday, but the quality was really really bad. I'll take another one on Monday with my actual camera (not my phone).

I've referenced this before, but you can see in this Mecanum-wheel drive that the rollers don't slip:
http://www.youtube.com/watch?v=vAiwLRGsNrE&hd=1

Originally Posted by LegoMindstormsmaniac

Thinking about the ways that slipping would show -- would it not be visible when being in a defensive battle with another robot? If so, have you all watched this video http://youtu.be/OCll0hrBmRc? Watch at ~1:15 for a pull test VS an all-omni robot. To me, it suggests that VEX Mecanum Wheels and VEX Omni Wheels are (for the most part) equal in a traction/slippage standard. Perhaps I am wrong in thinking that? Perhaps we need to conduct a test similar to this with some slightly heavier, higher-powered robots?
[...]

Thank you for bringing this up actually. The problem with this test is that both motors stall, so (as has been suggested before) the torque needs to be increased. If I have a chance I will run a similar test with more torque on the wheels.

This video refers to my second point (which is still a topic of debate I think) that Mecanum wheels only have 70% the traction of wheels without the angled rollers but the same coefficient of friction (omni wheels).


btw Jordan,
NXT or RCX?



Oh yeah and while we're on the topic of debates and arguing and whatnot, can we at least agree that the rollers do not turn when the robot is turning about the center of a square whose vertices are the 4 Mecanum wheels?

[Rich Kressly]

Here's a blog post with a picture of our "H-Drive" configuration from this part year.
http://robobraves.blogspot.com/2011/...75-and-78.html
As a VRC rookie, we used 5 motors in the drive, won our first event with the new/small omnis in place, then switched out to the 4" in omnis for increased speed and subsequently also won our second event.
Here's some video of it with the full robot:
http://www.youtube.com/watch?v=7pFvtm6WBPM
http://www.youtube.com/watch?v=CCzaqyqXH5I

We're early in our team brainstorm process for Sack Attack, but I'm sure we'll strongly consider using a similar drivetrain.

[RoboDesigners]

Originally Posted by dontworryaboutit

I tried to take a video on Thursday, but the quality was really really bad. I'll take another one on Monday with my actual camera (not my phone).

I've referenced this before, but you can see in this Mecanum-wheel drive that the rollers don't slip:
http://www.youtube.com/watch?v=vAiwLRGsNrE&hd=1

Ok. I guess I was wrong. :\

Originally Posted by dontworryaboutit

Oh yeah and while we're on the topic of debates and arguing and whatnot, can we at least agree that the rollers do not turn when the robot is turning about the center of a square whose vertices are the 4 Mecanum wheels?

Um... maybe. I don't think I want to get in another argument that I probably won't win...

//Andrew