pic: 2826 Halonomic Drive Prototype

[big1boom]

Nice mechanum drive. Looks awesome, with some pretty cool machining.



Are those custom made wheels?

[Andy Baker]

This is very, very cool. As with everyone else, I have a few questions:

What is the purpose of the scissor-like suspension? Is it to provide active suspension for the wheels, or does your wheelbase footprint need to change during your application?

What is the diameter of those wheels?

Maybe I am looking at this the wrong way, but shouldn't the wheels be swapped from front to back? Looking at the top view of the robot, I believe the mecanum roller positioning should look like this:

\........../
\........../
\........../
............
............
............
/..........\
/..........\
/..........\

This is a very inspirational mechanism. I would love to see what all is going on within each of those suspension members.

Sincerely,
Andy Baker

[EricH]

Originally Posted by Andy Baker

Maybe I am looking at this the wrong way, but shouldn't the wheels be swapped from front to back? Looking at the top view of the robot, I believe the mecanum roller positioning should look like this:

\........../
\........../
\........../
............
............
............
/..........\
/..........\
/..........\

Andy: It depends on the exact application. Mecanums work in either the X configuration or the O configuration. If you have an X on the top, you have an O on the bottom. The bottom side matters the most, naturally.

For the purposes of this discussion, X and O refer to the shape of ground contact. X looks like Andy's picture above; O looks like the following:

/.............\
/.............\
/.............\
...............
...............
...............
\............./
\............./
\............./

Their current configuration is an X pattern on the ground. This provides slightly easier turning. An O pattern would be just as valid, but is slightly (just slightly) worse at turning due to roller angles.

Either pattern works with the same programming, and it's fairly easy to try both. Just swap the wheels front-to-back or side-to-side. Or, in the case of this robot, flip the whole thing over.

[Jared Russell]

I think you have that backwards - an O pattern on the ground provides more efficient turning. In fact, I don't believe that a square robot with an X on the ground can turn at all.

To turn clockwise with an O pattern on the ground (or an X pattern looking at the top of the robot):
The front left wheel provides equal torque in the FORWARD and RIGHT directions.
The front right wheel provides equal torque in the BACKWARD and RIGHT directions.
The rear left wheel provides equal torque in the FORWARD and LEFT directions.
The rear right wheel provides equal torque in the BACKWARD and LEFT directions.

Hence 100% of the torque (for a square robot) is around the axis of rotation. For an elongated drive base with 45 degree offset rollers, the percentage is less than 100%.

Now consider a clockwise turn for the opposite configuration - X on the ground (looks like an O looking at the top of the robot): None of the four wheels can supply torque around the axis of rotation (only towards or away from it). For a non-square drive base this isn't the case, but the percentage of output torque that aids in rotation is still going to be less than 100%.

* All of this assumes 45 degree offset rollers with negligible friction. If the rollers are sufficiently sticky, you can achieve an increased turning effect with the X on the ground (at the cost of decreased strafing efficiency).

[EricH]

Originally Posted by Jared341

I think you have that backwards - an O pattern on the ground provides more efficient turning. In fact, I don't believe that a square robot with an X on the ground can turn at all.

To turn clockwise with an O pattern on the ground (or an X pattern looking at the top of the robot):
The front left wheel provides equal torque in the FORWARD and RIGHT directions.
The front right wheel provides equal torque in the BACKWARD and RIGHT directions.
The rear left wheel provides equal torque in the FORWARD and LEFT directions.
The rear right wheel provides equal torque in the BACKWARD and LEFT directions.

Hence 100% of the torque (for a square robot) is around the axis of rotation. For an elongated drive base with 45 degree offset rollers, the percentage is less than 100%.

Now consider a clockwise turn for the opposite configuration - X on the ground (looks like an O looking at the top of the robot): None of the four wheels can supply torque around the axis of rotation (only towards or away from it). For a non-square drive base this isn't the case, but the percentage of output torque that aids in rotation is still going to be less than 100%.

* All of this assumes 45 degree offset rollers with negligible friction. If the rollers are sufficiently sticky, you can achieve an increased turning effect with the X on the ground (at the cost of decreased strafing efficiency).

Are you sure about that torque pattern? You may be mistaking the contact pattern for the vector pattern.

Clockwise, O contact pattern on the ground, lines are vectors in a clockwise turn, arrows are vertical component of the vector:

\../ (^v)
/..\ (^v)

Clockwise, X contact pattern on the ground, lines are vectors in a clockwise turn, arrows are vertical component of the vector:

/..\ (^v)
\../ (^v)

So, the X pattern forms a circle on the ground with its force, while the O forms an x pattern for the same turn. Note that all vectors are in the direction perpendicular to the rollers for that wheel. If you have an O pattern, perfect square, the vectors go into the middle. An X pattern will go around the center.

Note that ALL patterns I reference are in relation to the ground contact. I don't care about what the pattern is off the ground.

Alan, the reason you may notice an O pattern turning better would probably be that the rollers are forced into the sides of the wheel, locking them temporarily, while an X pattern is easily spun due to the rollers not going into the sides. Assuming that the rollers were to stay in the center on all wheels (i.e., no friction with the sides of the wheel), I'm not sure that you'd see the same effect.

Again, if you don't like the results one way, swap end-to-end, side-to-side, or (in this case) top-to-bottom and see if you like those results better.

[Jared Russell]

Originally Posted by EricH

Clockwise, O contact pattern on the ground, lines are vectors in a clockwise turn, arrows are vertical component of the vector:

\../ (^v)
/..\ (^v)

Clockwise, X contact pattern on the ground, lines are vectors in a clockwise turn, arrows are vertical component of the vector:

/..\ (^v)
\../ (^v)

Eric,

The problem is not with the vertical components of the vectors but rather with the horizontal. Also, I think that what I call "X" and what you call "X" are different.

So, using your notation:

Orientation of torque vectors - Vertical components - Horizontal Components

What I called "O":
/..\ (^v)(>>)
\../ (^v)(<<)

No problem here. The vertical components want the robot to turn CW, and so do the horizontal.

What I called "X":
\../ (^v) (<<)
/..\ (^v) (>>)

Problem! The vertical components want to turn CW, while the horizontal components want to turn CCW. The two cancel each other out as long as the roller friction is negligible.

[AJ R]

That looks sweet! The wheels look nice, did you make them your self? From our experience with an omni drive, the building was the easy part, and the programming is the hard part. Ours was drivable, but difficult to control. Hope yours turns out better than ours.

[Tyler Olds]

Thank you for all the nice compliments, I will try to answer what I can, but this is a prototype and we have not gotten to test it on anything but a concrete floor. We are still making adjustments to the drive since we determined that we need to have an additional gear reduction so the motors do not kick out.

Originally Posted by Greg Needel

I like the concept alot, I sketched something similar for a project I had an idea for. Can you tell me how this works on grass? Also can you control the height of the main body by adjusting your struts on the fly or is it just a fixed suspension driven by a shock?

We have not tested it off road yet, but the hopes of the suspension that we have is to take it in different terrains. It is a fixed suspension, however we have holes drilled to adjust each leg.

Originally Posted by Chris is me

WOW that thing looks beastly! Are those 14 inch wheels? Got any video? That seems like it wond be really cool to see run and swerve. How much movement do the shocks have? Could they go, say, upside down? There's some small paralells to be drawn here; 1714 did holonomic in 2008 build, 2826 did crab in 2009, now those are reversed in the offseason... I guess I just find it funny.

It is a beast both in size and weight. I do not know the exact diameter of the wheels off hand. The shocks have a fully range of movement to the point that the frame will bottom out before the shocks give. The robot is not designed to drive upside down if it flipped over but it would be cool.

Originally Posted by big1boom

Nice mechanum drive. Looks awesome, with some pretty cool machining.
Are those custom made wheels?

They are custom made out of polyurethane that we made in a custom mold.

Originally Posted by Andy Baker

This is very, very cool. As with everyone else, I have a few questions:
What is the purpose of the scissor-like suspension? Is it to provide active suspension for the wheels, or does your wheelbase footprint need to change during your application?
What is the diameter of those wheels?
This is a very inspirational mechanism. I would love to see what all is going on within each of those suspension members.
Sincerely,
Andy Baker

Thanks Andy! The footprint does not need to change during usage, just to provide an active suspension. I should be able to post a couple pictures or video next week.

[Tyler Olds]

Got a short video uploaded on youtube. I took this at midnight on the spur of the moment so this is nothing professional. Look for a longer video in a couple weeks. There are still many more tweeks that will be done, hopefully sooner than later.

*Oh and the 2x4 was just randomly laying there*

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

[EricH]

Originally Posted by Jared341

Eric,

The problem is not with the vertical components of the vectors but rather with the horizontal. Also, I think that what I call "X" and what you call "X" are different.

So, using your notation:

Orientation of torque vectors - Vertical components - Horizontal Components

What I called "O":
/..\ (^v)(>>)
\../ (^v)(<<)

No problem here. The vertical components want the robot to turn CW, and so do the horizontal.

What I called "X":
\../ (^v) (<<)
/..\ (^v) (>>)

Problem! The vertical components want to turn CW, while the horizontal components want to turn CCW. The two cancel each other out as long as the roller friction is negligible.

I see now.

I also think that X and O should be "standardized" at a certain view for future reference. I'll let wiser heads than mine figure out whether you do it from the top or bottom, though.

[David Brinza]

Can this robot flip over and still drive? That would be awesome!!

[Alan Anderson]

Originally Posted by EricH

Mecanums work in either the X configuration or the O configuration....Their current configuration is an X pattern on the ground. This provides slightly easier turning. An O pattern would be just as valid, but is slightly (just slightly) worse at turning due to roller angles.

A true X configuration would have the wheels in a perfect square with the mecanum rollers all pointing to the same point in the center of the robot. Such a configuration cannot reliably turn in place under its own power, and cannot prevent itself from being spun due to external forces. With the wheels in a longer (or wider) rectangle, this problem is less pronounced, but it still exists.

An O configuration is noticeably superior in turning performance, with both forward and sideways performance remaining the same.