Slashdot Mirror
Autonomous Boats Will Be On the Market Sooner Than Self-Driving Cars (vice.com)
An anonymous reader quotes a report from Motherboard: In the autonomous revolution that is underway, nearly every transportation machine will eventually be self-driving. For cars, it's likely going to take decades before we see them operating freely, outside of test conditions. Some unmanned watercraft, on the other hand, may be at sea commercially before 2020. That's partly because automating all ships could generate a ridiculous amount of revenue. According to the United Nations, 90 percent of the world's trade is carried by sea and 10.3 billion tons of products were shipped in 2016. According to NOAA's National Ocean Service, ships transported $1.5 trillion worth of cargo through U.S. ports in 2016. The world's 325 or so deep-sea shipping companies have a combined revenue of $10 billion.
Startups and major firms like Rolls Royce are now looking to automate the seas and help maritime companies ease navigation, save fuel, improve safety, increase tonnage, and make more money. As it turns out, autonomous systems for boats aren't supremely different than those of cars, beyond a few key factors -- for instance, water is always moving while roads are not, and ships need at least a couple miles to redirect. Buffalo Automation, a startup in upstate New York that began at the University at Buffalo, just raised $900,000 to help commercialize its AutoMate system -- essentially a collection of sensors and cameras to help boats operate semi-autonomously. CEO Thiru Vikram said the company is working with three pilot partners, and intends to target cargo ships and recreational vessels first. Autonomous ships are an area of particular interest for the International Maritime Organization (IMO), which sets the standards for international waters. It launched a regulatory scoping exercise last year to analyze the impact of autonomous boats. By the time it wraps in 2020, market demand may make it so that we already have semi-autonomous and unmanned vessels at sea.
Startups and major firms like Rolls Royce are now looking to automate the seas and help maritime companies ease navigation, save fuel, improve safety, increase tonnage, and make more money. As it turns out, autonomous systems for boats aren't supremely different than those of cars, beyond a few key factors -- for instance, water is always moving while roads are not, and ships need at least a couple miles to redirect. Buffalo Automation, a startup in upstate New York that began at the University at Buffalo, just raised $900,000 to help commercialize its AutoMate system -- essentially a collection of sensors and cameras to help boats operate semi-autonomously. CEO Thiru Vikram said the company is working with three pilot partners, and intends to target cargo ships and recreational vessels first. Autonomous ships are an area of particular interest for the International Maritime Organization (IMO), which sets the standards for international waters. It launched a regulatory scoping exercise last year to analyze the impact of autonomous boats. By the time it wraps in 2020, market demand may make it so that we already have semi-autonomous and unmanned vessels at sea.
Boat and plane navigation is reasonably similar - it's 'head to waypoint', not 'navigate through twisty curves'.
That's partly because automating all ships could generate a ridiculous amount of revenue. According to the United Nations, 90 percent of the world's trade is carried by sea and 10.3 billion tons of products were shipped in 2016. According to NOAA's National Ocean Service, ships transported $1.5 trillion worth of cargo through U.S. ports in 2016. The world's 325 or so deep-sea shipping companies have a combined revenue of $10 billion.
Notice how none of these statistics address, at all, how much money there is in automating ships? Besides the hand-waving, the article doesn't address it at all.
I mean, I'm sure that there's some, but just because most cargo goes by sea doesn't necessarily mean anything in relation to whether automating ships can save any money or increase revenue.
How can we continue to believe in a just universe and freedom to eat crackers if we have no ale?
Maybe start with a pilot program.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
To use a car analogy, it's like a car that drives without needing a driver.
The environment is simply more forgiving and more constrained for boats, planes, space probes, mining equipment, farming equipment, tunnel service robots, and non-personal autonomous systems (aka pizza delivery bots on sidewalks). They all will appear on the market before a consumer-grade L5 car will be available to fleet operators on a large scale, meaning servicing everything from northern Alaska to southern Patagonia during any season in any city.
Everybody, including Google, has only demonstration systems that operate under very specific constrained environments. They are impressive (especially Google's system), push innovation (e.g., autopilot and supercruise), and are fun to think about, but nowhere near that they could start manufacturing them with a profit to be sold to fleet operators at Level 5.
These kinds of ships spend almost their entire voyage on "auto pilot" now, and they require a local pilot and tug to navigate into harbor. With the lack of personnel I'd bet insurance rates go down in case of piracy, the need for food and crew space goes away and can be used for more cargo. The ships will still be met and guided into any harbor by the same system we use now.
errr....umm...*whooosh* *whoosh* Is this thing on ?
A big container ship will have a crew of about 25. Of these, only 3 or 4 are directly involved in steering the ship: The captain, and a couple of deck officers, all of whom have other duties as well. And, as you said, they aren't paid much. So I don't see how this could possibly generate "ridiculous amounts of revenue" as claimed by TFA.
I reckon that fully automated ships ought to be the greatest boon to piracy since the invention of the cutlass. You don't even have to go out in potentially nasty weather to steal a shipload of containers -- just hack into the ship's network via any on board IOT device and run it up a remote beach where you can loot it at leisure while it's navigation gear reports back to the owners that it's en route to Montevideo..
You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
The problem is not that the technology is bad. The problem is that going to sea is much more complicated than non-seagoing people think. And, the sea is terribly unforgiving of any mistakes, incapacity or inattentiveness. We have gyrocompasses but we do check them against the magnetic compass. We have wonderful tracking radars with gyro stabilized displays. But, we still have lookouts for the things that radars do not pick up (like small boats). Satellite navigation provides us with accurate fixes, 24 hours a day. But, I still brought my sextant, and was expected to use it.
The engine room has a similar situation. Having engineers to maintain and repair the equipment is imperative. There are no repair crews when you are in the middle of the ocean.
A completely automated ship is even less likely than an automated airliner without a pilot.
Level 4 vehicles (fully self-driving, no human attendant, but geofenced to a specific area) are already here, working in real life, and have been all year. Hundreds of these are already ferrying the public around Phoenix, and they now have a licence for full commercial operation.
Expect to see thousands more real self-driving robotaxis in service in the 25 cities they're being tested in today.
Why would anyone engrave "Elbereth"?
They need to add lots of DRM to protect themselves from pirates.
Are these not being reported? Because global ship losses seem a lot less than one every other day
Why would anyone engrave "Elbereth"?
I'll add that most of the rest of the crew are there to fix stuff when it breaks while at sea. If costs could be lowered by reducing the number of crew and adding redundancy and shortening maintenance schedules so there's nearly zero chance of any critical systems failing during a 1 month trans-oceanic voyage, shipping lines would've done it already.
Ship crews are what they are because it's turned out to be cheaper to have ships staffed 24/7 by crew who can repair the exact item which breaks in-transit. This may not be obvious if you think of this from the standpoint of home or auto repair, where the cost of parts range from a few hours to a few days worth of labor. But on something as large as a ship, a part might cost several decades worth of a mariner's salary. And it ends up being cheaper to have someone aboard who can fix things, than to design all the systems to be redundant (add expensive backups) or swap out expensive working parts during maintenance because you're afraid they might fail during the next month-long voyage.
Of these, only 3 or 4 are directly involved in steering the ship: The captain, and a couple of deck officers, all of whom have other duties as well.
And the article mentions that the current systems are only semi-autonomous.
Means you won't be completely replacing the whole 3-4 guys steering the ship.
The captain will still be around, probably at least one of the deck officers, in order to overwatch the semi-autonomous system whenever it requires human supervision.
Compared to all the money involved in shipping cargo on huge container ship, the difference of salaries will barely register.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
I don't see how this could possibly generate "ridiculous amounts of revenue" as claimed by TFA.
Losses from pirates are high. Not only material losses, the cost of maintaining an anti-pirate army.
What the article doesn't mention is that robo-ships are allowed to use automated defenses. They're much more efficient killers than the minimum wage deck hands.
No sig today...
All those containers get loaded and unloaded by people operating cranes. Being able to optimize crane movement has got to be worth a lot of money in reducing the time a ship has to be at the dock.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
The navigation gear can not report a false position, as AIS etc. is picked up from orbit via sattelites ...
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Level 4 vehicles (fully self-driving, no human attendant, but geofenced to a specific area) are already here, working in real life, and have been all year.
From your link:
"On November 7, Waymo announced that it was going to start testing cars without a safety driver. "
So when did they start? That announcement and entire article says that they intend to start. It doesn't say that they have started.
Hundreds of these are already ferrying the public around Phoenix, and they now have a licence for full commercial operation.
You claim that they are already ferrying people, the article says that they intend to use the license to ferry people. Do you have another link? This one doesn't support your claim. What they intend to do and what they are currently doing are two different things.
I'm a minority race. Save your vitriol for white people.
There are no month long voyages anymore since about 100 years ... just saying.
Yes there are. It can take 2 weeks to almost a month just to cross the Pacific. Here's some Middle East/Horn of Africa times. That time includes several port stops, but just long enough to offload/load up cargo and resupply/refuel. Not long enough to do anything besides the simplest of repairs.
The only thing necessary for evil to triumph is for it to be pitted against a slightly greater evil
Losses from pirates are high. Not only material losses, the cost of maintaining an anti-pirate army.
Huh? Cargo vessels aren't allowed to be armed. There's the option of hiring maritime security, but for a shipping line to staff a security detachment on each each ship is prohibitively expensive, if not outright prohibited due to local laws at their ports of call. They could do it air marshal style with detachments on a small number of ships but that won't really work as a deterrent either. Ships can use passive or non-lethal defenses such as water hoses (which they will already have to fight onboard fires) or sound cannon. Somali pirates use small ships to board the cargo vessels so even the large wake and waves generated by the cargo vessel can be used defensively to make it impossible to board the ship.
Of course, we already have an "anti pirate army". There's the multinational maritime patrol off the Horn of Africa.
The only thing necessary for evil to triumph is for it to be pitted against a slightly greater evil