Go Down

Topic: Shelf Stocking Robot?? (Read 3137 times) previous topic - next topic

Johnny D

Does anyone know of a robot being developed that stocks shelves in stores?  I was thinking of it this afternoon and did a quick google search.  Might have been looking for the wrong thing though.  Any ideas?


They are called teanagers  ;)

Seriously, I think this a to complex task for a robot as there are to many different products in a store with way to many different packaging styles.

You have this kind of robots for warehouses which use standard sized pallets.


Nov 04, 2010, 01:46 pm Last Edit: Nov 04, 2010, 01:46 pm by Imahilus Reason: 1
as there are to many different products in a store with way to many different packaging styles.

I reckon this isn't too much of a problem.
The problem I see is... stores tend to change. Few stores have the exact same layout for years on end, which would make a robot economically viable.
If the robot is only going to do half upto three quarters of the shop, you still need your staff to stock the rest, which they often do when there are no or few customers.

All in all, its a true challenge.
So no, afraid I don't know any, though I'll eagerly check out anything that gets posted here, I'm curious as to how they solve(d) the various challenges!


The thing should have an optical system to see if some products have fallen or where they are in the shelf.  You Need to pull older products to the front (you don't want a product to stay there for ever)

Think this is way to complicated to make it an economical sollution.


This isn't exactly what you are looking for , but still pretty cool.


I saw a demo once by Kiva Systems.  These robots basically work by bringing the shelves to the box packer.  CVS decided to expand a warehouse they had with these robots in it and the robots did all the moving and  setting up in the new added on space!

Also, the sytem is completely dynamic, as in each shelve goes where it is most needed and doesn't have a home space.

Johnny D

Cool.  I've seen Kiva before, and seeing another video of them is what sparked the idea.  I was wondering if anything has come up on the other end.  

Yeah, there are a lot of challenges.  Especially since, as was pointed out, the different "terrains" change.  I'd imagine something with and RFID tag and a way to "see" shelf space, fix up items that are askew, check for mis-placed items, etc.  There's a lot going on there.

As far as being economical, I guess it would come down to the time it took to customize each product.  If it's an hour with an engineer to get the store setup and then 10 minutes of time with each new product, maybe it could become economical.  But yeah, there are a lot of challenges there.  

I wonder how much it costs a typical walmart to do restocking per year; if you could figure that out, youd have an idea if it's economically viable or not I think.  

Ran Talbott

Few stores have the exact same layout for years on end, which would make a robot economically viable.

That's not really a problem:  any robot sophisticated enough to load shelves will have enough processor power to carry re-loadable maps of the layout.

I think the biggest challenges are making the gripper flexible enough to pick up a huge variety of packages,  and making the vision system sophisticated enough to recognize when the shelves are "ordered" and "fully stocked".

The gripper(s) would need to perform tasks like lifting a heavy can of juice up and out of a shipping case,  then changing its grip so it could slide the can horizontally onto the shelf.

The vision system would have to recognize whether a customer had discovered a big sale on root beer and abandoned his Coke there instead of returning it to where he got it.  It would need to determine whether there's a big hollow space behind the cans neatly arrayed across the front of a shelf.  And deal with a lot of other situations that are easy for specially-evolved biological systems,  but very hard for designed-from-scratch mechanical ones.  I'm sure there are projects in various labs that can handle such challenges.  Possibly even a few systems being tried out in the field.  But I'm guessing they're still too spendy to deploy in a supermarket.

Johnny D

Yeah, that's a great point.  To properly check the shelves, you would have to have some sort of vision recognition:  not only for orderliness, but for properly placed items.  

For handling, I originally imagined a pneumatic sucker of some sort.  It would work pretty well for cans, boxes, etc.  Maybe not so much for milk, dog food, and other heavy items.  


any robot sophisticated enough to load shelves will have enough processor power to carry re-loadable maps of the layout.

Quite true, what I was thinking of... somebody needs to make the changes to the maps, allthough we can certainly dream of the robot doing that part too  ;D
Takes training to be able to make such changes, aswell as added difficulty for trying to get a feel of the store through a virtual map.

Apologies that I hadn't elaborated on that point.
PS: fun discussion.

Johnny D

I would think that with some wifi direction finders (I think I've read about such a thing) or maybe IR or other radio beacons in each corner of the store you could know pretty precisely where you are at any given point.

Also, I would imagine that some sort of RFID tag would be placed at each "destination" (IE, the Wheaties slot on the shelfs and the Diet Coke can slot).  So robot could go to where he thinks it should be in general with it's load, and then find precisely (say, RFID tags go in the upper left-hand corner of a grocery slot).  

The thing I can't get my head around is the imaging analysis.  What I mean by that is what's been mentioned a few times above.  One of these bad-boys has to:

- Recognize what's not supposed to be there and set it aside.
- Move everything that's supposed to be there into an orderly and presentable fashion.
- Add the new stock.

That probably requires it to scan the shelf and then do the recognition.  

I would imagine setting this thing up (ie programming for specific tasks) might just take a specialized technician a few day at the site:  take multiple pictures of each item for stocking (for image analysis I guess; and that's probably really standardized along with dimensions via the UPC and a barcode scanner), set out the RFID tags for each product with the positional reading, and observe a night of stocking.  Oh and measurements for the dimensions of the shelf being targeted (IE, how much can we get into one place).

Hmm.  Probably more complex than that. ;D

Go Up