Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Robotics

No Human Can Match This High-Speed Box-Unloading Robot Named After a Pickle (ieee.org) 94

Posted by msmash from the how-about-that dept.
schwit1 writes: Able to move 1,600 boxes per hour using just one arm, Dill relies on humans to keep it operating efficiently Pickle Robots says that Dill's approach to the box unloading task is unique in a couple of ways. First, it can handle messy trailers filled with a jumble of boxes of different shapes, colors, sizes, and weights. And second, from the get-go it's intended to work under human supervision, relying on people to step in and handle edge cases.

We asked Meyer how much Dill costs, and to our surprise, he gave us a candid answer: Depending on the configuration, the system can cost anywhere from $50-100k to deploy and about that same amount per year to operate. Meyer points out that you can't really compare the robot to a human (or humans) simply on speed, since with the robot, you don't have to worry about injuries or improper sorting of packages or training or turnover. While Pickle is currently working on several other configurations of robots for package handling, this particular truck unloading configuration will be shipping to customers next year.
This discussion has been archived. No new comments can be posted.

No Human Can Match This High-Speed Box-Unloading Robot Named After a Pickle More

No Human Can Match This High-Speed Box-Unloading Robot Named After a Pickle

Comments Filter:

  • Smokin! (Score:4, Funny)

    by DesertNomad ( 885798 ) on Monday April 19, 2021 @02:34PM (#61291440)

    I guess if I had to do a boring job like that, and at that rate, I'd be smoking too.

    • I think the smoke/steam is to show the robot can clearly work in non-ideal conditions. Too many companies do their testing in controlled, well-lit, clean rooms (*cough* Apple butterfly keyboard *cough*), this company wanted us to see they're not doing that.

      • cool.
        but can the pickle move to the next truck by itself.
        then repeat unloading boxes.
        also.
        when the lights go out.
        the pickle will continue unloading.
        just throwing it out there

        • but can the pickle move to the next truck by itself.

          Doesn't need to. It will sit in the loading bay and trucks will back up to it.

          • Also, the lights (electricity) goes out in a building far less frequently than they do in a person (sleep?). Just saying

          • maybe.
            but looking at the logistics.

            tractor rolls up with a trailer.
            then parks the trailer in a loading bay.
            the trailer maybe filled or not.
            tractor picks up a trailer and moves on.

            this robotic arm is stationary.
            so that means having an arm at every bay.
            bays are open air environments.
            dust and dirt and the elements get involved

            also.
            what happens when a box falls and spills its contents.

            also.
            what happens if a person walks up.
            and removes a box.
            maybe the box now goes missing

        • Re: (Score:2)

          by tragedy ( 27079 )

          but can the pickle move to the next truck by itself.

          It appears that the answer to that is probably "it depends". As it stands, this robot does appear to have a mobile base. Once in the truck, it can advance into the truck by itself. However, actually getting in and out of the truck is a question. Not all trucks magically mate perfectly with the loading dock so that a robot can just roll in or out of the truck. There may be a ramp or other system required. Something like that could be automated as part of an overall unloading system, but it's not clear from t

      • "you don't have to worry about injuries or improper sorting of packages or training or turnover." I am not sure how convincing an argument that would be to a CEO, considering there don't seem to be whole lot of companies out there that really care about their workers, or their customers. The only argument that will convince them is how many pink-slips they can hand out.

        • As a CEO of a company, I care about my workers and my customers. If I didn't, I don't think I'd have many workers, or customers. This is the basis of competition.

          Even if I didn't care about my workers, turnover is expensive. Even for something as basic as box unloading, there are costs associated with bringing someone onboard (getting them into your accounting system, making sure they have their documented training, etc. etc. etc.) as well as costs associated with their leaving, and finding a replacement.

          • The problem arises when a company grows so big that there are layers of middle managers and bean counters between the ones in power and the ones their power affects. Even if the ones in power have intact humanity, being isolated from their subjects reduces its effectiveness.

        • Re: (Score:2)

          by SirSlud ( 67381 )

          there isn't "care" and "don't care" there's a spectrum of caring from very little to a lot.

          but that's besides the point, you're responding to points that incur expenses to the employer, independently of why they would care whether or not a worker gets injured or quits so your kneejerk binary mode cynicism isn't even applicable to the point being made

        • Then make the injuries, occupational health and safety [osha.gov] regulations, and logistics arguments to human resources, legal, and shipping, and let them feed the arguments independently up to the C-levels. If they hear it a few times, they might start getting the idea, especially when it's laid out side-by-side with risk management.

      • Re: (Score:1)

        by baglunch ( 11210 )

        I was guessing it was to show that the clip is being run at real time, since you have a pretty good idea what steam dissipation looks like.

  • Working on a shipping dock is dangerous work and can take its toll on you physically. You can have all the safety in the world it will happen. Automating dangerous jobs is the appropriate choice, imo.

    • Re:It does take jobs, however (Score:5, Insightful)

      by cayenne8 ( 626475 ) on Monday April 19, 2021 @02:50PM (#61291486) Homepage Journal

      Working on a shipping dock is dangerous work and can take its toll on you physically. You can have all the safety in the world it will happen. Automating dangerous jobs is the appropriate choice, imo.

      You also don't have to pay all the people it replaces $15/hr.

      I think we're going to see this tech accelerate here quickly in the next couple years.

      • Re:It does take jobs, however (Score:4, Informative)

        by Hadlock ( 143607 ) on Monday April 19, 2021 @02:59PM (#61291520) Homepage Journal

        Also, they won't sue you when a box falls on them, robots never call in sick, you don't have to pay the robot's social security (employer half), unemployment tax (employer half), doesn't take smoke breaks, doesn't underperform when their girlfriend breaks up with them, never shows up to work hung over etc etc
         
        If your warehouse deals with a lot of the same widget, or truck loadout, then you can more reliably time how long it will take to load/unload a specific container, which is going to improve scheduling/freight throughput etc etc
         
        Warehouse workers are doomed here in 15-20 years for all but the very smallest operations where the warehouse manager is also the only person doing the loading/unloading, and even then probably part time.

        • Robots call in sick all the time. And then you need to repair them. Whereas you can always just hire another worker.

      • Kindof like self driving cars.
      • so I think the point you're hinting at (that the only way to save our lives is to throw ourselves at the mercy of our capitalist overlords and accept every decreasing wages) isn't going to save anyone here.

        Even Chinese workers with borderline slave labor couldn't compete with machines. There have been multiple stories of factories being told not to automate by the Chinese gov't in an effort to hold back social unrest (millions of densely packed unemployable people is a recipe for disaster).

        Raising w

    • Re: (Score:2)

      by fermion ( 181285 )
      Humans can work on docks under a variety of conditions. Humans may want climate control, but often wonâ(TM)t quit over it. Machines tend to break when they get uncomfortable.

    • Automating dangerous jobs is the appropriate choice, imo.

      Let's automate all the jobs we can. Ideally, we'd have robots doing all the work and sit around creating cool stuff and looking at other people's cool stuff. Whether that's code/electronics projects or art or 17-course meals.

  • That's still pretty tidy for a trailer (Score:1)

    by Anonymous Coward

    In their video the 'untidy reality' is still extremely tidy.

    Obviously these guys have never seen a FedEx or Hermes trailer :-) ;-)

    Also, all the boxes are exactly that, boxes. Flat and strong sides. In *reality* parcels aren't quite like that.

    • I wonder what the trailers of HDS [youtube.com] look like...

    • Re: (Score:3)

      by tragedy ( 27079 )

      Of particular note is the fact that this robot is for "unloading" boxes. In principle, a robot that can unload can load as well. So I wonder where they stand on that feature.

      • It could probably be extremely good at it if the computer running the thing is aware of all the incoming boxes, their dimensions, and the order in which they'll arrive. Tetris is a lot easier if you know in advance what the pieces are, can perhaps rearrange them or place a few into a holding area until you want them, and can think much faster than they move.

        • Re: (Score:2)

          by tragedy ( 27079 )

          I mean, there is always the danger that you stack the boxes too well and then they just vanish and the boxes on top drop down to the bottom. :)

          Seriously though, yes, it does seem like it should be able to do that quite well. Of course, it probably also helps to be able to define the center of gravity of the box, whether the contents can shift and to be able to estimate how much weight the box can bear. All of those are things a human being can do reasonably well. It really does seem to me that it should be

  • Cool machine, but this thing isn't... (Score:4, Insightful)

    by bjdevil66 ( 583941 ) on Monday April 19, 2021 @02:57PM (#61291512)

    ...eliminating human jobs yet.

    If you watch the entire video, a human comes along and takes the box the robot placed on the conveyer belt.

    And they still have to work out reaching WAY back into the truck. The arm isn't long enough yet and will need counterweights.

    BTW - Nobody should be using questionable cardboard for packaging because a machine like this would eat the cardboard for breakfast.

    • Re: (Score:1)

      by Anonymous Coward

      The boxes also seem to weigh nearly nothing (they look empty).
      I wonder how the pneumatic grabber would handle a box that's too heavy to pick up from one side, would it let go, or tear the box?

    • it's to eliminate some jobs. Let's say it eliminates 10% of warehouse jobs. 1st, you just saved 10% on labor, minus the cost of the machines. But 2nd, you just reduced the number of jobs in your industry 10% without reducing the number of laborers.

      Supply and demand works both ways. Supply stayed constant (same # of workers), demand fell (10% fewer jobs). What do you think happens to your costs as an employer (read: wages)?

      All of this is possible when you stop thinking of people as people and start t

    • Re: (Score:3)

      by jwdb ( 526327 )

      And they still have to work out reaching WAY back into the truck. The arm isn't long enough yet and will need counterweights.

      Already worked out. From TFA: "The system is mounted on a wheeled base, and after getting positioned at the back of a trailer by a human operator, it'll crawl forward by itself as it picks its way into the trailer. "

    • BTW - Nobody should be using questionable cardboard for packaging because a machine like this would eat the cardboard for breakfast.

      I think any machine would struggle to treat packages as badly as some delivery and package handling humans. I have had do not fold packages folded and stuffed into my mailbox, fragile packages that look like they have been used as footballs. When I pack anything for sending now I automatically assume it will be mistreated and pack accordingly with extra padding inside.

    • Re: (Score:2)

      by tragedy ( 27079 )

      If you watch the entire video, a human comes along and takes the box the robot placed on the conveyer belt.

      That's just for their demo. What happens to the box after it goes onto the conveyer can vary, obviously. It might be coming into a sorting center where a whole series of conveyors routes the boxes to different bins or areas where they get loaded into other trucks, or they get opened and the contents used, or they get loaded onto shelves. Or there could just be another robot or set of robots at the end of the conveyor that take the boxes and stack them on shelves, or humans could do that, etc., etc. It doesn

  • Named after a pickle.... (Score:3)

    by cygnusvis ( 6168614 ) on Monday April 19, 2021 @02:59PM (#61291518)
    Its name should have been RICK
  • Wages are still too low to make this robot profitable.

    • Re:Wages (Score:5, Interesting)

      by sdinfoserv ( 1793266 ) on Monday April 19, 2021 @03:12PM (#61291574)
      Loading dock wages are $19-$23 / hr ; with full benefits you're in the rage of $40-45/ hr employer cost. https://www.salary.com/researc... [salary.com] That's $100K per year per staff to eliminate human liability and error - no vacation, no sick days, no workman’s comp to juggle. At $300K each, these robots have a 3 year ROI.
      The next question is what do you do with the 10's of thousands of loaders who no longer can support their families with $50K wages and full benefits or the taxes these former employees no longer pay (because they have no jobs). To the corporations it doesn’t matter. Let’s just not talk about why homelessness, opioid use, mental health problems, gun violence are all up no jobs and the rich continue to get richer.
      • wages in my area are $9 for that work

      • Where did you come up with $300k each for the robots? TFS said $50k-$100k with another $100k in operating costs each year. One person could operate 5 of them and they could cover 3 shifts replacing something like 27 employees. (Twice as fast as humans, so 5 robots == 10 humans. But since they can run 24/7, those 5 robots can cover 3 shifts, just needing 1 human per shift to baby-sit.) That's $1.7m saved in year one if that is your situation, and future years it will be $2.2m anually.

        If that's an accurate pr

        • $300k was a theoretical number, out of the air, out of my arse, just a number to start from. Companies don't charge manufacturing + %, they charge what they perceive is the "value" included all the warm fuzzie marketing and branding to maximize profit. Sure, they can add al lthe stuff you mention and charge and easy million.

          • The builders of the Dill robot might not have enough manufacturing capacity yet - so they will sell to companies paying more than the "ticket" price first (i.e. we can sell you this robot for $75k late next year, for for $100k early next year).
            Also, there might be some "growing pains" in the first version, i.e. more downtime than estimated, or higher repair costs, or "work speed" lower than quoted - all could add up to higher initial investment.

      • Let's just not talk about why homelessness, opioid use, mental health problems, gun violence are all up no jobs and the rich continue to get richer.

        The rich can try and not talk about the very problem they're creating. But rest assured they're not putting just putting deadbeats and derelicts permanently out of work.

        Best case scenario? The uprising only results in a little violence and a lot of taxes and fees to pay for all the unemployable.

        Worst case scenario? The rich will find themselves at dinner. Served as the main course.

      • you radicalize them into militias and use them to overthrow the democratically elected government, installing yourself as Dictator for Life.

      • "The next question is what do you do with the 10's of thousands of loaders who no longer can support their families [...]"

        Nope, that question has already been answered by very smart people: UBI.

        Cost for goods is going way down, money for work too. The remaining difference will have to be paid off as dividend of overall wealth as an UBI for everybody, if Utopia shall not collapse.

        And we won't want it to collapse.

        I make 60k in a bullshit job that will be gone in 5-10 years and play online with a buddy who is

    • Not really. It will replace multiple people and people cost more than wages. There are taxes, insurance, training, onboarding costs, offboarding costs, mandatory breaks, overtime, and many more. Basically, it will pay for itself if it replaces 5 people who are working for $15.00 and assuming a 25% overhead.
  • mabye dill can aslo sort our recycling.

  • I am sure it makes sense for certain types of businesses, and hopefully it can also eventually load trucks as well, but for a single-purpose solution it seems terribly inefficient— it just unloads bulk packed trucks to a conveyor, and apparently doesn’t do any sorting.

    In comparison, if it is only twice as efficient as a human, how much improvement would you have just with an automatic following conveyor? How much gain would you get with a magic carpet on the truck, or bins for the boxes?

    Cool to

  • If you give the robot a task to do in a typical (or any) environment and let it take care of part of the work within an operating range+safety margin, it can call people in when necessary [marshallbrain.com] to handle what it can't. This could (in theory) apply to printed [nytimes.com] or handwritten [wikipedia.org] text/speech/handwriting [wikipedia.org] recognition as well.

    You have to wonder if robotic replacement will happen fastest when robots start working alongside people and slowly start absorbing the corner cases upon observation -to- minor improvements -to- ri

  • Fish, plankton, sea greens... protein from the sea!

  • TFA says 1600-1800 boxes/hour vs. 800 "for a single human in top form". But a human can't keep that up for an 8-hour shift, and a robot doesn't take lunch breaks or even need to stop to pee in a juice bottle. So the robot moves as many boxes per full shift as three humans.

    Pick a middle number of $75k to deploy, amortized over 5 years = $15k/year, and $75k/year to operate. That's $90k/year. So it competes favorably with humans which cost $30k/year.

    At $7.25/hour humans might be cheaper. But $15/hour is $

    • Why would it be $75k to operate? Assuming a highly skilled tech to program and maintain? If so he probably will operate10 of these in parallel. Likely have one per truck bay and 10 truck bays doesnt sound excessive. How many do you think amazon has? Whats the math if you had 3 techs and 20 robots? Bet you cant even hire illegals that cheap.

      • "Why would it be $75k to operate?"
        Electricity cost. Maintenance. Prepared downtime (seen as a cost instead of a percent of time when the device is being maintained, fixed, ...). Replacement of worn parts (cost of new parts and work to install them). Transport to and from the factory/repair center for things that can't be replaced locally.
        Human supervision costs in time taken to move the robot from trailer to trailer, and in time taken to sort through "red flag" issues - cases when the robot finds a situatio

    • But a human can't keep that up for an 8-hour shift, and a robot doesn't take lunch breaks or even need to stop to pee in a juice bottle. So the robot moves as many boxes per full shift as three humans.

      A human is also limited to an 8 hr shift. The robot isn't.

  • The jobs these displace are physically damaging (humans aren't designed for that kind of repetitive motion which is destructive to joints and discs) and should be eliminated to the greatest practical extent.

    "Lumpers" in these bottom-tier break-bulk jobs are paid little and would be better off under UBI. Have some logistics history:

    http://blog.drivekandj.com/tru... [drivekandj.com]

  • Lifting boxes from the side and top? (Score:4)

    by swilver ( 617741 ) on Monday April 19, 2021 @04:36PM (#61291910)

    Don't know what is in those boxes, but a box of decent weight is not gonna like being pulled at from the side or suction lifted from the top. Cardboard will rip if not well supported from the bottom if the box has any kind of weight to it.

    • Re: (Score:2)

      by Tom ( 822 )

      My thought exactly. Some of the boxes appeared to have some weight to them, so depending on what it is that you're shipping/receiving, it may be ok - but a lot of possible packages will refuse to cooperate.

  • The factory workers start calling it Dildo?

  • "you can't really compare the robot to a human..."

    Said Peter Piper, CEO and master pickler at a robotic company named after a tart cucumber.

    Names like this make you want to Google your way to a toilet where you'll Tweet about your projectile vomiting.

    Guess you're in a hell of a pickle when the robot goes down, eh?

    (The robot, will see me out now.)

  • https://youtu.be/E-IwGbEIU08 [youtu.be]

    Thought this was going to be something new and exciting. Looks like just another robot arm :/

  • Post-scarcity economy.
    Nice. Like it. Material Utopia is arriving.
    Very, very weird having more and more machines doing the bulk of useful work but nice none-the-less.

    Now can we please have the boxes normed and reusable. And the stuff in them built to last and repairable.
    The ecosystem is taking toll and we're in irreversible deep trouble if we don't stop and fix this.
    This will be the next big challenge of human civilization.

  • ... in addition to the other benefits, it won't claim that you treated it differently based on the color of its paint job.

  • Now it need OCR to read "This side up" and keep that side up - unlike most human package handlers who read it and flip the damn thing.

    • Careful! If it looks for things like that all it takes is arrows pointing at different sides to create a Captain Kirk-style paradox in which it blows up.

Slashdot Top Deals

In the long run, every program becomes rococco, and then rubble. -- Alan Perlis

Close