Rolls-Royce's Vision of Shore-based Control Centers for Unmanned Cargo Ships | Hacker News Reader

Rolls-Royce's Vision of Shore-based Control Centers for Unmanned Cargo Ships

gcaptain.com

45 pointsprotomyth44 comments

It's hard for me to believe that the crew is the high-cost component in oceanic shipping. From what I've seen they're mostly from the Philippines and similar low-wage nations. Besides, getting rid of them isn't just a question of steering and throttling. Most of the crew is engaged in maintenance for most of the journey. Are the nerds on the holodeck somehow managing that as well?

How much maintenance is required at sea due to attempts at stretching out maintenance normally performed in dock?

If you can remove the crew cost, does that free up enough capital to perform regular maintenance of various elements to reduce failures at sea?

You are then left with losses due to structural damage at sea, which then disappears in the accounting books as self-insurable. If a ship takes a few extra days to get to destination because it was sailing around storms, there is no incremental cost beyond fuel (and if using wind/solar, the fuel cost disappears too).

I have to read that in my internal naive MBA's voice. After enough "optimization", one is left picking up pennies in front of the steamroller. Many things can go wrong at sea. Many of those things will send one's ship and all cargo aboard to the depths. Insurers will not embrace the technology TFA describes.

Is it cheaper to have an emergency maintenance team on call 24/7 in a few key locations and just fly them out when there's an issue, or do the events tend to happen too fast to save the ships from a few hours away?

Some shipping routes pass through some fairly remote stretches of ocean. While it's true that few issues would require instant attention, just getting attention within twelve hours might be difficult. And now a widely-dispersed fleet of large seagoing helicopters and their support infrastructure is a link in the chain. Here I thought ship maintenance was a challenge! Perhaps the biggest problem is that when maintenance is really a priority, severe storms are likely to be involved. Flying into those in order to land personnel and equipment on a troubled ship in order to perform maintenance should not be a regular event in oceanic shipping.

This sort of innovation might catch on for less remote routes first, however.

Lol. Good luck landing a maintenance crew plus tools on a ship that's dead in the water in the middle of the Pacific Ocean. The available helicopters don't have enough range.

No, but I submitted an article a bit earlier about the first drone delivery to a ship. It does occur to me that the same technology being developed for remote surgery could be developed for remote ship repair.

Drones don't have any more range or cargo capacity than piloted helicopters so that's not going to solve anything. No one is going to risk sending a drone to land human technicians on a disabled ship. We are decades away from being able to build remotely operated repair robots that could begin to approach the capability and reliability of human technicians. Building a surgical robot that stays fixed in one operating room and moves a limited set of tools a few inches is orders of magnitude easier than building a robot that can roam around a huge ship and repair the hull, engine, electrical systems, painting, etc.

I would expect the droids on board to be built into the ship at strategic places. Since this whole automated thing won't happen for a bit, I would bet drones will improve and might just drop parts depending on the quality of the bots. If they build automated ships, I would expect they would be built to have automated repairs.

The crew isn't just there to fix stuff that breaks. They're performing routine preventative maintenance constantly as the vessel is underway. That's in addition to performing inspections that would be difficult to do remotely, and training to handle emergencies that need to be dealt with _right now_ (you're not going to flying a response team out to a ship that has caught fire in the middle of the ocean... at least not if you expect to recover anything... that sort of thing needs to be dealt with by people on board).

Crew cost is expense, not capital.

Well-spotted, actually maintenance is as well.

> How much maintenance is required at sea due to attempts at stretching out maintenance normally performed in dock?

I don't follow... How does removing the crew make it any more worthwhile to perform maintenance while in port? The issue with trying to handle all maintenance when the ship is "parked" is that the ship only makes money when it is working. By performing routine maintenance while the ship is under way, the owner reduces their opportunity cost pretty substantially (far more than they would save by getting rid of the crew).

Additionally, "a few extra days" can represent hundreds of thousands of dollars for a large container ship. The economics of container shipping depends very much on getting from point A to point B as quickly as possible, and then spending as little time at point B as you can, before continuing on to point C, and so on. A ship sitting around in port is an expensive hole in the water.

Perhaps it's not about direct crew costs, but the additional security needed to keep an onboard crew safe[0]? Without humans aboard, hijacking may become far less of a risk, since the ship doesn't need to be human-pilotable at all. With that resistance to hijacking, not to mention the fact that no crew lives would be at stake in a piracy event, the shippers may be able to pay less for insurance and security services.

[0]: http://www.theigc.org/project/the-economic-costs-of-piracy/

That's not much of a factor. Piracy is only a risk in 3 regions right now, and even there the risk is low.

I'd guess that the marginal efficiency of cutting a few crew positions isn't much, but no longer needing to support human life at sea would seem to cut significant overhead.

As soon as you need one person for a mission-critical task at sea, you need a bunch of other stuff (including people) on the ship just to support him. A medic, a cook, food, water, shower, toilet, transportation (the ship's boats), maintenance on all of the above, etc.

It doesn't have to be the high-cost component to be worth automating. Seems like you could program a route by satellite without too much difficulty and just watch telemetry from there. If something goes really wrong, just have the ship sit in the water until a roving maintenance crew arrives. Unlike flight or driving there is a pretty wide margin for error before safety consequences kick in (unless it's near shore).

That doesn't work in the real world. A disabled ship won't just obediently sit there like a broken-down truck on the side of a highway. A ship without power will drift with the current and crash into a reef, or simply break apart and sink when it gets caught in a storm and can't keep its bow into the weather. http://gcaptain.com/tag/el-faro/

Okay, but how much of a difference does it make to fly a crew out to the ship only when needed vs. having the crew on board?

That wouldn't work either. Current helicopters lack the range to cover many Ocean shipping routes. And they can't safely land on a disabled ship in the middle of a severe storm.

This is the most sensible justification I've seen yet, but I note that for some time, ships have gotten larger and larger. To torture a metaphor, when all our eggs are in one basket, we want somebody watching the hen-house. It seems common that while small systems are designed for maintenance-free reliability, really big systems worth lots of money are designed to require occasional human intervention. Maybe the observation robot can't maneuver well enough to notice coolant leaking out of the oil pump (I'm not a maritime diesel mechanic, but let's imagine that these oil pumps get hot enough to require their own cooling systems), or maybe it's not equipped to tighten the relevant fitting. When the remote operator gets on the phone with the manufacturer to complain about having to shut down, they're not going to have much sympathy for her when she relates that there's no one on-board to turn a wrench. In the meantime, the shipper is paying pro-rated late fees while the ship floats in the Northern Pacific without an engine.

This would kill seafaring unions. Don't underestimate how appealing this is to shipping companies.

Note: I am not arguing against seafaring unions, just explaining the reasoning of corporates (having worked in head office of a shipping company).

They are not quite on the same level of stevedore unions, but they are immensely powerful and militant. The term "sea lawyer" [0], which predates unions, adequately describes them. They know their rights, are completely inflexible and prone to strike.

Simply not having to think about union demands is enough to make a CEO salivate over autonomous ships.

I like sailors, I especially like sea captains and know a few captains of huge container ships. I have major respect for them. But I completely understand why corporates would end seafaring unions is they have half the chance.

[0] http://www.dictionary.com/browse/sea-lawyer

I think that it is too simplistic to think that all Rolls-Royce will do is replace the onboard crew with computers and an on-shore control center. Granted that I don't know anything about shipping I can see some areas that could change:

- Not having a crew onboard means a ship can be underway any amount of time as long as it doesn't run out of fuel or misses cargo deadlines (I don't know but there's probably a good amount of stuff that is allowed to take a good while to arrive).

- Not having a crew onboard means that the drawbacks for having many smaller ships is significantly less than it would be otherwise.

- Not having a crew onboard means onboard maintenance either needs to be automated (not very likely to be possible) or it should no longer be needed. Perhaps smaller ships with replaceable parts or sufficient backup systems are low maintenance enough to only need servicing at ports.

Usually running costs far outweigh the cost of materials so shifting the balance to having less of the former and more of the latter makes complete sense to me. Perhaps I'm missing key information but I can definitely see why Rolls-Royce would want to go in this direction.

I see a lot of comments about what happens when something goes wrong with helicopters not being able to reach the ship if something happens. But I fail to see how that is different from the current situation with crew on board; surely they can't fix every problem that could occur. And in the event of an emergency the crew could require immediate evacuation which a drone would never need. Doesn't seem like a problem?

Gah, a gCaptain article! There goes my evening...

Seriously, gCaptain is probably my favorite i-know-nothing-and-this-is-awesome industry blog. Would be fascinating to know more about (and perhaps work on!) the state of the art in maritime technology.

Agreed! Anyone aware of any other similar blogs focused on otherwise mysterious industries?

I find http://russianforces.org/ fascinating -- the author is an expert on russian nuclear weapons, and writes about the development of new missiles and subs.

http://alert5.com is a pretty good aggregator for aviation news.

On a total aside, I wonder what VFX house they got to do this, because it all looks great. Usually these random corporate concept videos all look really cheesy.

How would they enforce security? Remember that pirates exist, would the savings on human captains be enough that they could sail around dangerous areas?

I know this is a planned concept, but surely security is a major concern. Can anyone with further shipping industry knowledge chime in?

I had the same thought. But maybe, it's less of a concern with an automated ship.

If the steering and engine rooms are sufficiently hardened, that is. It's one thing for someone to get onboard, but it's an entirely different thing to stop the boat. Without humans onboard, who will stop it? And, without a crew, there aren't any hostages. If you're remotely controlling the boat, you could just keep going and bring the pirates with you to the nearest maritime/port authority (Coast Guard, etc...). There is still a risk to the cargo, but how are pirates going to unload a container ship that is still moving in the ocean with an uncooperative captain controlling it remotely?

They can just damage the ship. It would be also ridiculous to think that there won't be an onboard manual override for controls.

I'm not saying that they couldn't. I expect that they would. But, for large container ships, isn't the bigger concern hostages? No crew = no hostages.

I mean, pirates aren't going to be able to dock a large container ship at any random port. And even if there is a manual override, why wouldn't it require a key of some kind?

I guess the point I'm trying to make is that the threat model changes when you don't have a crew on board that you can threaten. Some parts of piracy (access) may be easier, but others more difficult.

> How would they enforce security?

Make them submarines?

Might that also lead to reduced stresses on the vessel, and less chance of something breaking? A failure might mean the whole lot ending up on the bottom of the sea?

You missed the joke. It's obviously not economically viable to build huge cargo submarines.

The Russians were thinking about doing just that with their SSBN sub design. It didn't come to anything, but people have been doing the research in a serious manner.

If there's no crew on board then there's no risk to human life or ransom to pay, so the consequences of a pirate attack are lessened.

I doubt unarmed or lightly armed crews are much of a deterrent to piracy anyway.

> If there's no crew on board then there's no risk to human life or ransom to pay, so the consequences of a pirate attack are lessened.

I don't think this is really true. Each container is worth $20k - $1M,

https://people.hofstra.edu/geotrans/eng/ch3en/conc3en/table_...

so the whole ship of ~10k containers

https://en.wikipedia.org/wiki/Container_ship

would be worth $200M - $1B or more. With 20 or less crew on board, and lives being valued even in the (expensive) US at only ~$10M, it's very likely that the value of the cargo is much higher than the crew. Not to mention the cost of the boat itself, which can exceed $100M.

Those numbers seem to be confirmed by this talk

http://static1.1.sqspcdn.com/static/f/1340097/17625392/13341...

This is more along the lines of what I was thinking. As "we all know", security isn't security, it's merely a case of "how long do we have?". Pirates do go after the contents of ships, that's a fact. I have nothing to cite for the following, but I can't imagine that the pirates actually want people on board. To a business - unfortunately - people are replaceable for much less than what their cargo costs.

360 degree long-range turrets and sticking to international water seem like the best strategy they have here. :^)

This is an awesome concept video.

Autonomous sailing, which sounds like a natural next step after self-driving,autopilot etc would be easier, but probably more points of failure and less chance of repair.

I definitely don't know the complexities of shipping vessel navigation/maintenance but it seems like a somewhat quick way to try this out could be try this with motors/sensors and have repair crews ready in ports (on standby) and just airlift them to ship, in case of failure.

This reminds me of the "Da Vinci" virus form Hackers movie, which threatens to capsize an oil tanker fleet. Who would have thought that movie was so visionary!

Hackers will now be able to remotely hijack ships to Somalia.

Through FBI back doors.

Does anyone else think the "captain's chair" looks like it came out of a Star Trek movie?

It also reminds me of the chairs in Beer's Project Cybersyn.

https://en.wikipedia.org/wiki/Project_Cybersyn

Reply on news.ycombinator.com