Live data from Hacker News

Autonomous Ships

bloomberg.com

11–20 of 74 posts

Re: Autonomous Ships

#11

As guess this would qualify as great news for sea-faring pirates.

Pirates are after hostages, not cargo, generally. It's very hard to crack the cargo anyways - it's usually stored in a solid steel 20-to-40 foot shipping container with solid steel lashings bolting everything together. You'd need to be able to cut through a lot of steel, but then moving the actual cargo itself would require, well, a large vessel. Selling it would be a very involved process. All of this puts sea piracy of drone vessels well outside of the scope of a group of armed men in a motorboat.

Potentially you could also establish measures that make it hard for a human to get aboard the ship, such as no walkable spaces, or tall fences, now that you no longer need to accommodate human crew.

Re: Autonomous Ships

#12

By one consultant's estimate, moreover, carrying sailors accounts for 44 percent of a ship's costs. I find this hard to believe even if autonomous ships can be done without a bridge. The Emma Maersk has 15,000 TEU and a crew of 13 http://www.emma-maersk.com/specification/

A crew that has to be paid, fed, insured, provided habitable accommodations on the ship including sleep, cooking, and personal hygiene, and provided benefits. I'm sure that's not cheap (though perhaps not 44%, either).

Re: Autonomous Ships

#13

As guess this would qualify as great news for sea-faring pirates.

Pirates are after hostages, not cargo, generally. It's very hard to crack the cargo anyways - it's usually stored in a solid steel 20-to-40 foot shipping container with solid steel lashings bolting everything together. You'd need to be able to cut through a lot of steel, but then moving the actual cargo itself would require, well, a large vessel. Selling it would be a very involved process. All of this puts sea pirac…

And no crew that you could threaten to obtain the keys to move freely.

Also, an armed rescuing force would have no hostages to worry about.

Re: Autonomous Ships

#14

By one consultant's estimate, moreover, carrying sailors accounts for 44 percent of a ship's costs. I find this hard to believe even if autonomous ships can be done without a bridge. The Emma Maersk has 15,000 TEU and a crew of 13 http://www.emma-maersk.com/specification/

"Crew costs of $3,299 a day account for about 44 percent of total operating expenses for a large container ship, according to Moore Stephens LLP, an industry accountant and consultant."[1]

The Maersk triple E's have a crew of 13 and consume ~136 tonnes/day of bunker fuel[2] which costs currently ~$315/tonne[3], for a daily fuel cost of $42,840. So right off the bat that number is very suspect. Even for other ships like the CSCL Globe with a crew of 31, that's a 12x difference.

I don't understand how that estimate can possibly be true. Ship captains and engineers make ~80k annual and crewmen make under 40k. For a crew of 20-30, $3,299 would be basically bang on for a daily crew cost but there's no fucking way fuel and depreciation only cost $4,200 a day. Even a 4000 TEU ship (tiny) at 17 knots (snail) burns $15,000+ a day.[4]

[1] https://www.bloomberg.com/news/articles/2014-02-25/rolls-roy...

[2] http://www.scdigest.com/ontarget/13-09-12-1.php?cid=7401

[3] http://www.tsacarriers.org/calc_bunker.html

[4] https://people.hofstra.edu/geotrans/eng/ch8en/conc8en/fuel_c...

Re: Autonomous Ships

#15
post #6
post #2

As far as my knowledge goes about the shipping industry: ships are already 90% automated (since they simply go straight forward for days/weeks). There is not much to do for a captain/officer whilst in the middle of an ocean. AFAIK most of the humans on board of a vessel are there to perform maintenance (most of which can be done during transit). Yes, we can automate the last 10% to make the ship sail itself, but we'd…

Maybe fully autonomous ships should have maintenance performing robots then?

That is a way harder problem than automated sailing.

Also a way harder problem than just designing in longer service intervals.

Re: Autonomous Ships

#16
I would love to work on this kind of software.

It's exactly the type of things I find interesting, it's a shame I didn't make different career choices when I was younger but it is what it is I guess.

Re: Autonomous Ships

#17
post #6
post #2

As far as my knowledge goes about the shipping industry: ships are already 90% automated (since they simply go straight forward for days/weeks). There is not much to do for a captain/officer whilst in the middle of an ocean. AFAIK most of the humans on board of a vessel are there to perform maintenance (most of which can be done during transit). Yes, we can automate the last 10% to make the ship sail itself, but we'd…

Maybe fully autonomous ships should have maintenance performing robots then?

That or have a small service vessel that sails with a school of autonomous ships handling maintenance.

Re: Autonomous Ships

#18
post #13

Earlier quoted context omitted.

Pirates are after hostages, not cargo, generally. It's very hard to crack the cargo anyways - it's usually stored in a solid steel 20-to-40 foot shipping container with solid steel lashings bolting everything together. You'd need to be able to cut through a lot of steel, but then moving the actual cargo itself would require, well, a large vessel. Selling it would be a very involved process. All of this puts sea pirac…

And no crew that you could threaten to obtain the keys to move freely. Also, an armed rescuing force would have no hostages to worry about.

No one to threaten to make the ship stop either. There's a fair chance a would-be pirate would end up in Hamburg.

Re: Autonomous Ships

#19
"The U.S. Coast Guard has estimated that human error accounts for up to 96 percent of all marine casualties."

But if we somehow manage to account for 90% of all eventualities when programming, how are we going to be safer?

Post reply on HN