Automate the Freight: Drones Across the Sea

When you think about which of the many technological advances of the 20th century had the most impact on the global economy, which one would you rank as the most important? Would it be the space program, which gave rise to advances in everything from communications satellites to advanced composite materials? Or would it be the related aerospace industry, which stitched the world together so tightly that you can be almost anywhere on the planet within 24 hours? Or perhaps it’s the Internet, the global platform for buying almost anything from almost anyone.

Those are all important, but for the most economically impactful technology of the 20th century, I’d posit that the lowly shipping container and the containerized cargo industry that grew around it win, hands down.

How could an almost technology-free steel box compete with the bells and whistles of spacecraft, jet airliners, and a global network of computers? When you think about it, moving goods from point A to point B is one of the key tasks in a global economy. And when your globe’s surface is 70% water, moving goods by ship is something that you need to be really good at. Until the mid-1950s, almost every ship was loaded by hand, with boxes and crates winched from dockside up into holds, where it was arranged by stevedores who as often as not helped themselves to whatever they wanted. Ships took weeks to load, cargoes were relatively small, and shipping was expensive.

By Photocapy (Flickr) [CC BY-SA 2.0], via Wikimedia Commons
The standardized intermodal shipping container changed all that — cargo was now fairly secure and handled in bulk by cranes. Ships could be loaded and unloaded quickly, turning weeks dockside to hours. Ships have become enormous as a result, and the reason you can order a widget on Ali Express or eBay and have it show up pretty quickly is because it got crammed into a container in China and crossed an ocean along with thousands of other of these lowly steel boxes, each full of something someone is convinced they want.

Land, Air, and Sea

In my earlier article about automating freight, I focused mainly on land-based, long-haul freight shipped by driverless trucks. But I also touched on automating the ships that ply our oceans. Most container ships these days have barely a dozen crewmembers aboard. Technology has automated away most of the jobs on a ship, and little stands in the way of complete automation.

I have little doubt that day will come, but there’s a problem with this mode of freight: ships are slow. The modern container fleet averages about 15 knots, meaning that a crossing of the Pacific can take something like three weeks. That’s an amazingly short trip compared to the middle of the last century, but it might be too long for some kinds of shipments — produce, for instance. Yes, you can ship across the ocean by standard air freight, but at a high premium compared to surface shipping.

Could there be another way? San Jose, CA-based startup Natilus thinks so, and they’re working on autonomous freight aircraft to ply the same routes that container ships currently dominate. They have ambitious plans: 200-foot long UAVs that will tote 100 tons of freight across the Pacific in 30 hours or less. The company and the concept appear to be in their infancies now, but they plan a test of a 30-foot scale model of their freighter this summer in San Francisco.

But in this day and age of self-crashing cars and fears of drone shootdowns, what makes Natilus think they’ll be allowed to fly a drone the size of a Boeing 777 above population centers? Here’s the clever bit: they won’t. Natilus intends to fly their drones completely over the ocean. What’s more, the drones won’t even use airports; they’ll be seaplanes, and will land, unload, load, and take off at or near seaports. There’s obviously going to be an efficiency hit compared to container ships, since cargo will need to be handled more than once. But if Natilus figures out how to leverage the venerable 20-foot shipping container format, my guess is the loss of efficiency will be more than covered by a 2000% faster transit time.

Obviously, Natilus and its eventual competitors have a huge number of problems to solve. Surprisingly, I think the drone part of the equation isn’t one of them — we’ve already got a pretty good idea how to make big UAVs and fly them safely over land. I think the problems lie more with the infrastructure that’ll need to be in place on both ends of the journey. Seaplane landings are no trivial matter, and even without passengers to fill up the “For Discomfort” bags on a rough landing, the cargo and the plane itself will still require some pretty smooth water to use as a runway. There will also need to be automated barges to ferry cargo to and from the docks, facilities both at sea and on land to build, and a thousand regulatory minefields to cross.

As the saying goes, whatever the laws of physics don’t specifically prohibit is just an engineering problem, and in this case, I’ll bet that economic forces will overcome the technical issues and provide us with much more affordable overnight overseas deliveries. And if it happens, it’ll be due to innovative thinking and automating away the problems.

105 thoughts on “Automate the Freight: Drones Across the Sea

  1. Autonomous oceangoing cargo vessels were been done (experimentally) decades ago. From friends in the maritime industry, the discussion then usually turns to the need for people on board to (1.) fix/alter/accommodate the autonomous stuff when it goes wrong (2.) take the blame/sign the papers in port (ie. the entirety of international maritime law) and (3.) deal with stuff the programmers couldn’t foresee from numeric data – which is a lot. So, there are a few people on board still, and they’re often kind of busy – particularly the ships’ engineers.

    So, aircraft. Aircraft will fly themselves now (autoland/auto takeoff are already a thing as is autopilot). What isn’t a thing is automating a last minute taxiway change, a wind shift that changes runways and thus requires a lot of runway crossing clearances etc. at the last minute in a busy freight hub, the failure of the ILS equipment at the destination. And oh, yeah…#1 through #3 as above.

    Why do we have “drones” in the air now? Most of these are UAVs and there’s a difference; a dedicated UAV-only runway in some dusty place and a dutiful human in a trailer somewhere flying the thing. This is the more likely outcome of such a venture…provided you can get rid of the rest of the traffic at a freight hub.

      1. The guy is right. In the grand scheme of the cost of the ship and cargo just a couple of humans can run the show. And the rest can be automated.

        The world isn’t driven by engineers. Its economics and i think that’s OK.

  2. The Soviets thought giant ekrano planes would revolutionize transport decades ago, Boeing revisited the idea after them & everyone comes to the same conclusion, it’s not worth it.
    I doubt automation changes the math that much.

    1. It does look pretty awesome, doesn’t it? ‘Course, I think it’s all marketing fluff and the real thing will look much more like a conventional plane – in fact, scroll down to the “careers” section to see someone working on a model with does look like a conventional plane. But we can hope!

        1. Good point, guess that’s just a “these are the sort of folks we’re looking for” sort of picture. Still, pretty renders are all good and well, but I’ll believe the reality of it when I see it…

    1. Or the whole ‘drones hacking into other drones’ issue. Better make sure it is resistant to spoofed GPS signals. :/

      I’m pessimistic about this. There will be sky pirates no doubt about it.

  3. How could an almost technology-free steel box compete with the bells and whistles of spacecraft, jet airliners, and a global network of computers?

    It’s not about how fancy the solution is, it’s about how appropriate it is.

    1. Yes, and not only that, the real technology behind the simple shipping container is the methods used in the background to track each container, so they get to the right place on time. That “global network on computers” is an integral part of why the shipping container is such a success.

      1. Oh interesting, I didn’t know that. So there’s more technological wizardry to the success of shipping containers than the simple idea of a standardised corrugated metal box.

  4. “and the reason you can order a widget on Ali Express or eBay and have it show up pretty quickly is because it got crammed into a container in China and crossed an ocean along with thousands of other of these lowly steel boxes, ”

    TIL “Pretty quickly” = ~ 2 months.

    Seriously anything you’re getting inside 2 or 3 weeks ain’t bobbing across the ocean in a container.

    1. I have ordered a few bits and bobs from China, complete with the caveat that it could take a couple of months to show up, and then I see it in a week or ten days. I suspect that some of those items get loaded into containers, and then just before they dock in LA or Oakland, they send an email to an agent who repackages them and sends them out to me out on the Great Plains.

  5. On the hack perspective, this might be a new skillset requirement by the likes of the Somali pirates who often board and hijack commercial freighters and tankers. Attempting to take over a ship that can’t be easily controlled manually could reduce the number of maritime losses.

      1. It’s hard to hit a fast moving target with an unguided, shoulder-launched anti-tank rocket. You’d probably be better off building an off-the-shelf RC plane and flying it ahead of the drone so it gets sucked into an intake.

        1. First, its over water! So you have a aircraft in flames hitting the water 2-300 mph. There won’t be much to salvage, except for some floating debris among a backdrop of flaming fuel. I’m also not sure how much you’d expect to get for a bunch of waterlogged “Tickle me Elmo’s’ anyway.

          Now, Hacking a drone and commanding it to land so you can raid the contents might work, but would probably work only as a plot on NCIS LA. And given the tech industry claimed it was reeling from the immigration ban from Somalia, proves it is a realist threat, given the quality of their engineers.

  6. I’m more interested in the current generation of dirigible aircraft.The Natilus site doesn’t say how much tonnage it can carry in on flight (that I can find), but the Lockheed prototype dirigible can carry 20 tons in one shot.

    1. And the Paul R Tregurtha can carry 68k long tons. Ships built to seaway max carry 28kt.
      Unless all you do is ship fresh produce or emergency supplies around the world it just doesn’t make sense.

        1. The legend lives on from the Chippewa on down
          Of the big lake they called ‘gitche gumee’
          The lake, it is said, never gives up her dead
          When the skies of November turn gloomy
          With a load of iron ore twenty-six thousand tons more
          Than the Edmund Fitzgerald weighed empty
          That good ship and crew was a bone to be chewed
          When the gales of November came early

        2. The Tregurtha is the largest container ship on the planet, and seaway max is by definition ships built to operate places other than but including, the great lakes. Locks on the great lakes are larger than those on the st Lawrence seaway.
          My point being that 20t is 3 orders of magnitude smaller than even the comparatively small seaway max container ships which are another third smaller than the largest ships in use today.
          Air cargo cannot compete in terms of tonnage or efficiency, only speed, which for many things does not matter. Shipping speed is not the limiting factor in car sales or refining raw materials.

  7. Ever hear of Amdahl’s law? This is just a silicon valley fantasy to fleece the gullible venture capitalist. The cost of crew is minor relative to the cost of fuel and engines. There were several 747 freighters sitting at an Asian airport for over a year, abandoned by the owners because the engines were timed out. The airframes are almost worthless.

  8. Myself, I’m a big advocate of dirigibles & navigating by drifting in the Jetstream ( different directions at various altitudes) saves much fuel &, when you are above the clouds, solar charge top-ups are available. Splitting the cloud water gives you lift-juice. No crew means no altitude sickness.

    That said, robust uplink security would be an issue & I would imagine governments will panic at the prospect of gigatonnes of material arriving from asia.

  9. I guess I dont understand the point of making it autonomous – Transporting cargo by air is more expensive due to a significant portion of the plane’s fuel and thrust being used just to keep the thing in the air. Ships on water don’t have to use energy to stay afloat, but plowing through the water is slower. Back to the planes though – making them autonomous would only save maybe a salary or two of crew on the plane. The major cost of flying a Boeing 777 jam packed with cargo across an ocean is the fuel, not the 2-4 person crew it takes to fly it.

    Everyone just loves the word “drone” and while autonomous stuff is actually really cool, its not going to bend basic principals of science and economics

    1. They could fly it more slowly as you don’t have to base trip time around the pilot’s endurance (or put two crews in it with one sleeping while the other one flies). Their website indicated they expected it to fly across the Pacific in 30 hours vs 11 for a existing cargo planes. So that would cut down the fuel needed to overcome aerodynamic drag.

      1. True, but the fuel cost per ton of cargo is still many orders of magnitude higher for air transport vs ship. If the planes gained twice the fuel efficiency and there was absolutely no crew, air shipping might become notably cheaper. It would however, still be orders of magnitude more expensive than via ship.

        1. True, but that might be a net positive value proposition for someone. The guy who started FedEx got laughed out of the room by his Yale business prof when he presented the idea for overnight delivery. “Why would anyone pay for that?” Never underestimate the power of impatience.

          Ans, “Why would anyone ever need more than 64k of RAM?”

        2. And don’t think Boeing or some of the major air transport companies like UPS or FedEx haven’t thought about it. If it was viable I think they would be doing it. But they aren’t.

          1. They might have thought about it, but they can’t do it, big companies have big momentum, it even takes them years to know they have a problem. You could say IBM was reasonably quick in responding to the Apple threat, compared to UPS responding to the FedEx threat (12 years). The corporate bureaucracy just won’t let them do it. Xerox for example, had the 1990s in the palm of their hand at Palo Alto, but could not do shit about it. Fedex is too big now, Boeing is too big, another 5 years maybe they’ll either buy Space X or begin competing. Some good portion of this momentum/stagnancy is that their shareholders will sue them if they’re not focussing on their “traditional” markets, it’s not a joke or cynicism that it’s all about quarterly earnings, because it’s all about quarterly earnings.

            So, it’s up to the disruptors to blaze the trail, even if it has been eminently do-able for 20 years, and there’d been 10 companies whose business interests it would further, they are kept on too tight a rein by shareholder expectations and upper management overcautiousness.

          2. Like RW said…

            If you want to understand the “psychology” of large business and their reaction to change, read the “Innovators Dilemma” series of books (the “Lean Startup” is in the same basic ecosystem).

      2. Modern jet engines operate at optimal fuel efficiency at subsonic speeds and altitudes of 30 to 40K feet.
        Flying them lower and slower likely would increase fuel consumption since it would be fighting against thicker air.

        1. That’s just the way the fans and compressors are sized. The design parameters are, “We want to fly fast, because we don’t want to carry twice the water, and catering supplies, plus sleeping berths, and double the crew when flights last as long as Lindbergh’s solo, so what’s the more efficient way to do that?”

          Because of that, there are numerous anti efficiency adaptions, like a heavy pressurisable cabin, and larger wing area than might be required in thicker air. Other adaptions to higher speed flight will make them less efficient at lower speeds and altitudes, it doesn’t mean efficiency at those altitudes cannot be done.

          Another trap for the lazy of thought is that cD is the same at all speeds and air densities, no, it’s true for a narrow Reynolds number range, outside of which odd things can happen. The trap leads to thoughts such as “If a super low cD is possible at 200mph 6,000 feet, then surely aircraft flying at 40,000 and 600mph should be that exact shape too.” and vice versa.

  10. Faster stinkbug hideaways and things that would die on long trips as invasive species. Tit for tat, we (US) just gave corn blight to Africa. Can’t wait for some of the willies from downunda.

    1. Aussie here – don’t think we have much that would be very invasive. Except the shit we imported (deliberately, at times) from Europe…

      Poisonous, on the other hand, yeah, we’ve got lots of them. Could be a new industry – export funnel web spiders for free, then export the antivenom for great profit.

  11. That rendering is rubbish from a graphics and an engineering point of view , I’ve seen a lot of aircraft over the years (I’m from a three generation aviation family) and that design is just wrong. So what are the chances of a company being so aeronautically clueless and still viable? None. And their website does not work at all without scripting.

    Speaking of the laws of physics, if you are going to push something through the air the velocity to travel distance to weight to volume ratios are very important to the question of if the energy required makes the exercise economically viable, and that is assuming you have enough room on board for the fuel required.

    Move on, nothing to see there…

  12. I see a use case for these drones, but it would be about as far from container based shipping as you can get.
    1. The fuel efficiency is orders of magnitude worse than conventional shipping.
    2. The drones will probably not be able to standard steel cargo containers.
    3. There is no existing infrastructure to provide the right mix of air/sea ports.

    This implies that the expected use cases for these drones would be situations where none of the above disadvantages actually matter:
    1. Fuel efficiency doesn’t matter if the cargo is extremely valuable, or illegal. In these circumstances, delivery time trumps efficiency.
    2. High value commodities will be shipped in smaller amounts, to limit loss.
    3. The lack of regulated infrastructure is an advantage if your cargo is illegal.

    1. 1. Fuel efficiency, ie cost, is something you compromise with speed. Same reason people fly to go on holiday.
      2. Well that’s kindof the whole point, so I think they’d design them specifically to work with existing cargo containers.
      3. They don’t need airports, just sea ports. Or really, just somewhere to bring the cargo to, then bring it across the water to where the thing’s floating. Probably would have some sort of floating harbour with a cargo terminal at one end, and the plane at the other, complete with refuelling facilities etc.

      Existing shipping is slow, so it’s made profitable by sending HUGE quantities. This wouldn’t be a vast-bulk system, it’d be something between existing air freight and shipping. Closer to air freight, but using shipping support facilities to make it cheaper. It wouldn’t carry such huge amounts of cargo, so it wouldn’t need the vast crane networks and robot transporters etc that sea cargo ports have.

      If it lives or dies, I think it’ll be for economic reasons. There might be some technical thing that kills it stone dead, but I can’t think of any obvious one. Seaplanes worked fine back in the day before airports were common. This is just a seaplane but with no pilot.

      That said, is it worthwhile eliminating the pilot? Is that a large part of the cost of flying freight? Would you need to have fully qualified pilots, if they’re not using ordinary airports or flying over land? Is there any international maritime airspace law?

    2. As fair as illegal cargo goes, though, these are a bit conspicuous, no? You still have to get your drugs onto dry land where people can buy them. Bloody great unregistered drone landing by the coast is going to draw a bit of attention to the boats that come out and unload it.

      Drugs don’t go off, so there’s no need to ship them quickly. They can wait. The only important factor with shipping drugs is getting them past the law. And even then it’s a scattergun kind of business, you ship through enough that you can write off the losses. Actually since you bought the gear for next to nothing in Pakistan or Colombia, it’s no great loss, it only becomes valuable once it’s landed in the country you’re going to sell it in.

      So for shipping drugs this wouldn’t really help. Other illegal cargo, I dunno. The same things I’ve said here also apply to guns. What else is there? Shipping humans like that would be a bit… horrible.

  13. Shipping containers are one of my favourite pieces of technology too.

    Whatever happened to the idea of using kites/sails to speed up cargo ships? Seems like a more environmentally sustainable option than aircraft.

    1. It’s got jet engines, you’ll have a job catching it in a Cessna. Maybe a Learjet. Even then it’s not like they ship diamonds and Mona Lisas in steel shipping containers, you might get a shipment of posh handbags if you’re lucky. Then you’d need somewhere to land the thing, and a way of unloading it all before it sinks because you haven’t got the proper facilities to land it at.

      You’d need a lot of investment in jet planes and electronics just to get up to the thing in the first place. I dunno if most Somali sea pirates can even read and write.

      You might get script kiddies, in the latest craze since swatting, hacking into their controls and crashing them into the house of the kid who called them gay over Counter-Strike (or whatever kids play these days).

  14. “…he cargo and the plane itself will still require some pretty smooth water to use as a runway.”

    Never flown in a seaplane, have you? Float planes can land in pretty rough water. Flying boat style planes are capable of landing in even rough water, and here size matters taking it up even further.

  15. What a crock. Planes fly at 30-40k feet and slightly subsonic because it’s the most fuel-efficient flight profile. Going low and slow is nice for BBQ, but stupid for an aircraft that you want to operate efficiently.

    The cost is not the crew, it’s all that kerosene required to lift a hundred tons into the air, so there isn’t really a meaningful saving to be had in automation.

    This is clearly a bunch of people who don’t understand any one of freight, aircraft design/operation, or business/economics. They just want someone to fund them to build a really big toy plane.

    1. For efficiency at low altitudes they’d need to go with props and a different wing then what they have.
      Another problem with the design the wing it does have is set vary far back so the aircraft would be very nose heavy.
      It would need to be a large strait wing vs a delta and for sea plane use it would need to be high mounted.
      A working concept probably would end up looking a lot like a Martin JRM Mars or a Consolidated PBY Catalina vs what is shown.

  16. “little stands in the way of complete automation.”

    Well apart from Weather, Marine Safety, Piracy, Terrorism, Marine and International treaties, Marine Standards(which believe me can often be traced back to Nelson time)….

  17. There is a reason seaplanes aren’t around anymore. Loading/unloading logistics are just a plain nightmare. Even WW2 as soon as land based airstrips in the Pacific became available flying boats like the Mars and the Spruce Goose lost all efficiency.

    There’s also the problem about them not being economically viable in terms of fuel cost as stated above. Mass cargo is mostly about making it as cheap as possible per ton-mile. And the best way to do that is bulk. As many containers as possible together on a massive ship. This entire idea goes counter to what the market has gravitated towards over the past decades.

    To me this presentation and idea seems just a little too slick. It feels like VC bait.

    1. Wasn’t the Spruce Goose Howard Hughes’s answer to a lack of metal available in WW2? So he built a giganormous wooden seaplane that could barely stay in the air?

      You know Hughes’s OCD is the reason Vegas is so big now? Up until a certain point, corporations weren’t allowed to own casinos, individuals had to. And they had to go to whichever office to sign the documents. By this point Hughes’s mind was stashed in a filthy mason jar somewhere in the corner, so he was NOT going to leave his room and go out into the filthy filthy world below.

      So he made some phone calls. He got the law changed so that he could buy his casino hotels through his company. Which he could do over the phone, through his middlemen. This opened the doors to corporations generally, who could raise huge amounts of investment money, to build the deity-offending monoliths you see now.

    2. Another factor aircraft ranges have improved considerably after the end of WWII allowing them to cross oceans in one go.
      Modern jets can fly directly from LA to Tokyo and if a flight does need a stop over it can land in Hawaii.
      A cargo drone would be able to do the same though it would be easier to just make a crewed aircraft for the task as you won’t need to get any laws changed and it would be able to use existing airports.

  18. I just say two words: “startup” and “San Francisco”.
    (Well, actually it’s three words; but who cares, i bet this sort of creative counting makes perfect sense for the people involved or invested in this kind of startup culture…)

  19. Natilus needs to beat this. Seriously—

    The C-5M has a take-off capacity of 270 000 lbs, flying 2150 NM, with a crew of three.
    Negotiable items: re-engineing is always an option; range and take-off weight are inversely proportional; and the crew of three is a military requirement; it is highly doubtful that the third crew member, a ‘loadmaster’, would be required except in the terminal area, ie, for loading and unloading on the ground.

  20. While an interesting idea, I think it’s more likely that we’ll get tunnels connecting the north america to asia which will have a constant stream of autonomous trucks flowing through them. It would be faster than harbor freight and have a reduced fuel cost. A properly organized network of recharge/battery swap stations could further reduce the cost by storing solar power in batteries on site.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s