Picking Up Upside-down Totes?

[Deucalion]

For the past couple of days I have been hearing mutterings from my team (STEMpunk 4531), as well as from many here on Chief Delphi concerning the difficulty of picking up an upside-down tote as compared to one found rightside-up. So to put the speculation to rest I ran a few experiments 10-12 times each, and determined quite reliably that at a contact pressure of no less than 6 pounds, any arbitrary end-effector should be able to grip the sides of an upside-down tote so long as the friction coefficient between the surface of said end-effector and the plastic walls of the tote is made equal to or greater than the coefficient between plastic and a pair of sweaty boy hands.

I didn't even need a calculator for this one, but if anyone else has a spread sheet or a graph or something showing exactly what it takes to grip a plastic bin at varying distances and with different materials, I would certainly like to see it, even if I wouldn't necessarily make one myself.

Now, of course there are other problems involved with upside-down totes aside from simply picking them up. For example; How do we flip them right side up for stacking quickly and efficiently, why not stack them all upside down, and how do we get our robot to recognize and accept both orientations in auto or even tele-op are all tricky issues and I would love to hear what you guys have to say about them!

[nixiebunny]

We figured that a tote-flipper would be more sensible, since a right-side-up tote is a lot easier to deal with in the robot. An arm that grabs the far top edge of the tote and pulls it towards the bot while driving backwards would achieve the job.

Actually, we figured that it's probably not necessary for playing the game pretty well. It might be needed to win CMP, though.

[Steven Smith]

Thanks for testing and making your first post to Chief Delphi a good one

Could you describe how you measured the 6 pounds of force?

[Michael Hill]

Do you really need to? There's enough that are right side up. Perhaps it would be easier to just move the upside down ones out of the way.

[rich2202]

If there were only 2 possible orientations of the Tote, then only stacking like totes is an easy solution.

But, you still need to figure out what to do with totes laying on their side.

[Deucalion]

Originally Posted by Steven Smith

Thanks for testing and making your first post to Chief Delphi a good one

Could you describe how you measured the 6 pounds of force?

Honestly, it was a complete guess and the although I did actually run the "experiments" 3 different ways 10-12 times, the entire post was laden with dry humor and mild sarcasm. Really, that was just what it felt like to me as I picked them up and set them down, BUT I did go and check the friction coefficient between skin and plastic, and it looks to be ~0.98 as described here -> http://hypertextbook.com/facts/2005/skin.shtml for dry skin. Given this and the weight of the totes at 7.8 pounds you can deduce that to grip the sides with dry hands you would need about 7.9 pounds of contact pressure, maybe a bit less with sweaty hands, so I wasn't too far off the first time when I said "no less than 6 pounds"!

[MrForbes]

If only we could find a COTS source for human skin!

(looks like we'll have to settle for phone grippy things)

[protoserge]

A tote flipper is not at all a complex or challenging thing to make. Look at Stack attack for ideas

[Koko Ed]

Originally Posted by stinglikeabee

A tote flipper is not at all a complex or challenging thing to make. Look at Stack attack for ideas

Stack Attack was more about knocking totes down than stacking them up. The human players did most of the stacking.

[Deucalion]

Oh, and what about picking up the recycling bins, if you ask me, anything that can pick up one of those should also be able to deal with the totes pretty easily.

[SamCyanide]

What I would do is have two pads (possibly rubber or something else that can grip plastic well) that are built off of arms with motors on them. This way, you can grab the tote and flip it right side up by using the motors. You could even do this in testing and have it automatically rotate the motors (and as a result the container) only a certain number of degrees.

[ehochstein]

'Snow Problem spoke a little bit about upside-down totes in this video. We made a couple of observations about how they move on the ground with zip ties.

[protoserge]

Originally Posted by Koko Ed

Stack Attack was more about knocking totes down than stacking them up. The human players did most of the stacking.

That's not true for all teams, but we'll focus on this game.

[nxtmonkeys]

Originally Posted by rich2202

But, you still need to figure out what to do with totes laying on their side.

You could just have a little arm with a spatula-like thingie on the end of it.

Here; I made a picture:
PICTURE
(yes, I am a bit bored. How did you know?)

[Koko Ed]

Originally Posted by stinglikeabee

That's not true for all teams, but we'll focus on this game.

The only robot I saw in 2003 who's sole focus was making stacks of bins was FRC 73 (a very ideal robot design for this year) and they got quickly figured out by having a bot assigned to stay close by and either force them to hold onto their stack of boxes or slap it down soon as it was released.
Teams realized it was much more worthwhile being King of the Hill than it was making easily destroyed stacks unless you brokered a deal to have them left alone.