Sailing Robots To Attempt Atlantic Crossing

Roland Piquepaille writes "The Times of London reports that seven robotic craft will compete in a race across the Atlantic Ocean in October 2008. One of them, 'Pinta the robot sailing boat,' has been designed at Aberystwyth University in Wales. Pinta is expected to sail for three months at a maximum speed of four knots (about 7.4 kph). Its designers hope the Pinta will become the first robot to cross an ocean using only wind power. This 150-kilogram sailing robot costs only $4,900. The transatlantic race will start between September 29 and October 5, 2008 from Portugal. The winner will be the first boat to reach a finishing line between the northern tip of St. Lucia and the southern tip of Martinique in the Caribbean. Here are additional details and links."

Comments

how about something a bit simpler

[jacquesm]

Like a robot that builds a house or so. A bit more useful too...

Robotics challenges are usually somehow tied to military objectives such as navigating a certain terrain, rarely do they focus on something constructive and creative.

Oh wait, another RP post...

Re:how about something a bit simpler

[Anonymous Coward]

St. Lucia and the southern tip of Martinique in the Caribbean Don't you see? They are going to corner the no-frills transfer to exotic holiday destinations?

Re:how about something a bit simpler

[BadAnalogyGuy]

Military applications can be considered "creative destruction", so it's not all mindless stuff.

On top of that, if you consider the current role of the army as a nation builder, then it is also important that the military be creatively constructive.

Re:how about something a bit simpler

[MagdJTK]

I'm no expert on robotics, but surely building a house is surely far harder than crossing the Atlantic for a robot?

Building a house requires all sorts of considerations about the land beneath it and requires a number of different skills.

Crossing the Atlantic requires going in a straight line for as long as possible.

Re:how about something a bit simpler

[HateBreeder]

It's all about money, right?

So if the defense department or the military will sponsor this, then its most likely to be something of use to them.

I think you should complain to construction or realestate companies,for not putting money into robotics.

The good part is that these things advance the state of robotics and will make a house building robot a little bit easier to design.

Re:how about something a bit simpler

[MichaelSmith]

I'm no expert on robotics, but surely building a house is surely far harder than crossing the Atlantic for a robot?

You should google around for fractal robots. Imagine lego blocks which are also robots. You broadcast a plan to the blocks, perhaps directly from a CAD desktop, and they self assemble into the intended object.

And yes, it is a little bit harder than crossing the Atlantic. But much more interesting (to me, anyway).

Re:how about something a bit simpler

[D-Cypell]

Building a house easier for a robot than crossing the atlantic? I have my doubts about that, even if you mean 'low grade' housing for use in the third world. Also, if a robot fails a sinks halfway across the atlantic, a few students get disappointed. If a robot fails, and the house it built a few days earlier falls and kills the family living inside, the implications are orders of magnitude more severe.

Also, I do see robotic ocean crossing as something useful and productive, but in addition, bear in mind that it is often the component parts that make real advancements in challenges like this. Power technology, navigation technology etc. Often the actual goal is secondary.

Re:how about something a bit simpler

[Anonymous Coward]

You really think the military cares about sailing robots that take 3 months to cross the Atlantic?

I think they can afford engines, and can spend their oodles of tax payer dollars on making them quiet, not to mention a sail tends to be quite visible...

And how is building a robot to do something that people actually find inconvenient or dangerous to do less useful than building a robot to do something that people have been doing for thousands of years?

Re:how about something a bit simpler

[peragrin]

you have to be able to know where the house is going to be built first.

Most of these competitions end up with learning remote guidance. This type of tech is what will allow planes to land themselves if something goes wrong with the pilot. Will Allow a ship to return to harbor on it's own if something happens to the crew.

All of this is relatively new technology. Sure radio controlled planes are 70 years old, but it has only been in the past decade that a camera could be fitted onto them. The tech needs massive amounts of refinement for practical purposes.

Re:how about something a bit simpler

[hey!]

Well, I'd guess that a robot house builder would be more complicated than the best available non-robotic option: prefabrication in a factory. So rather than having some kind of super-flexible machine that can do wiring, plumbing, framing and finish carpentry, you just set up stations with simple machines and conventional robots.

In any case, if you want to make houses by robot, you're going to have to constrain the designs and materials used around robot-friendliness. Once you've done that, you might as well use non-conventional designs that don't require complex craftsmanship or expensive materials (e.g., various adobe-like building methods).

Robotic sailing, on the other hand, is quite useful. It provides the ability to collect information over a large area and extended time at low cost and risk to human life. It may have utility in a future where energy prices make international shipping more expensive.

In any case, what I'd like to see is a robot that can replicate itself. The self reproducing 3D printer discussed here a few weeks ago is interesting, but too constrained; it can only make rather small parts, which accounts for its peculiar rod and fastener construction. A machine capable of actually creating and assembling its own mechanical pars would not only have to be capable of producing a wider variety of useful parts, it would eliminate the need for human expertise and intervention in making machines. That would break the duplication barrier for machines in the way that computers break the duplication barrier for information.

Re:how about something a bit simpler

[moosesocks]

Crossing the Atlantic requires going in a straight line for as long as possible. Not if you're sailing against the wind for at least part of the time.

You've also got to account for obstacles (admittedly not many) and currents (which could be very significant for such a small boat).

Re:how about something a bit simpler

[mshmgi]

Crossing the Atlantic requires going in a straight line for as long as possible.

I think only somebody who has spent a fair amount of time at the helm of a sailboat can truly appreciate just how complicated this is.

Winds vary to a great extent ... waves knock your bow from side-to-side (especially in a small craft, which this apparently is) ... currents can take you miles off course. The first two conditions can frequently require extremely quick and accurate responses to avoid capsizing - not so much w/ the currents of course.

Having an unmanned craft sail from Portugal to the Caribbean is more complicated than landing an unmanned craft on Mars. Once you leave the Earth's atmosphere, it's pretty much smooth sailing the whole way.

Re:how about something a bit simpler

[e2d2]

How could we ever use this technology for non military purposes? Well let's see - most of the world's food supplies are delivered via the ocean. How's that for a start?

Also, since you have ideas for better robots, why don't you get off your ass and build one yourself?

Seriously, this "meh, I could've done better" post is very typical, yet very arrogant.

Re:how about something a bit simpler

[GooberToo]

If a robot fails, and the house it built a few days earlier falls and kills the family living inside, the implications are orders of magnitude more severe.

Sounds like you've watched the SciFi channel too much. Robot builders does not mean lack of supervision. Nor does use of robots mean lack of general inspections. Frankly, human construction workers typically do piss-poor jobs in the first place until you are talking about high-end customer builders. For track homes, quality often barely able to pass inspection, and that assumes it is actually inspected. Chances are far higher even a poor robotic worker will increase quality a factor of ten or more.

Don't forget, your typical track builder is using illegal workers, workers that are drunk, workers that are high, workers that are more involved in dealing out the back of their truck, if they show up at all, or workers that dropped out of school in the fifth grade. Often, the hardest part of getting a house built, is simply getting workers to show up on time, sober, and ready to work.

Re:how about something a bit simpler

[CastrTroy]

I really don't see the problem with robots building houses. Most new houses look the same anyway. There's entire suburbs where all the houses look the same, and all the suburbs look the same as the other suburbs. There isn't much variation going on lately. Unless you look at expensive designer homes. Besides the actual cost of building a house is quite low. The real expensive part is the land. You can't get robots to create more land. Also, robots probably wouldn't build a house any better than humans will. Any problems with construction are more due to the use of bad materials, or only doing things "to code", and not going anywhere beyond that, and have nothing to do with actual workmanship.

Re:how about something a bit simpler

[drinkypoo]

How could we ever use this technology for non military purposes? Well let's see - most of the world's food supplies are delivered via the ocean. How's that for a start?

Most of the world's food supplies are transported on container ships, which are too large to move by sail - although there is at least one company selling helper sails which are basically big chute/kites, and which only work with a following wind.

It might be an effective strategy for cargoes which need to arrive quickly, but don't actually need to arrive at all - I'm thinking military payloads. Because then you can basically just launch a fleet of torpedo-equipped passive sonar-searching sailboats and let them do their own thing. But the simple truth is that there is an upper limit on the useful size of a sail (restricted largely by materials technology) and the oceans are getting MORE unpredictable as the weather becomes more unpredictable as we add heat to the system, so small craft are less and less likely to arrive at all.

In other words, this is a fun project, but it is not economically useful. Having a ship that can sail itself would be cool, but if a problem comes up, you still have to know what you're doing, and let's face it, current autopilots combined with manual control are not especially arduous in most situations.

Re:how about something a bit simpler

[Amouth]

i love to sail.. and while i have done it for years i wouldn't even consider taking this trip on something shorter than 90ft single or 50ft cat.. (if only i could be given the chance) this little thing is going to get thrown around like crazy.. i will be amazed if it makes it, and go forbid it runs into a depression..

Re:how about something a bit simpler

[FooAtWFU]

I really don't see the problem with robots building houses. Have you ever been involved in a) robotics or b) house-building? Combining robots' limited capacity for detecting and dealing with "messy" and unpredictable situations with the reality of putting things together outside of pristine, well-supervised factory conditions is not going to work out all that well, at least with the current levels of technology.

Besides, for the $1 million (+interest, maintenance, fuel) your house-building robot costs you, you can purchase an awful lot of labor that will do the same thing.

Re:how about something a bit simpler

[ColdWetDog]

It's all about money, right?

Anything about sailing is about money:

Definition of sailing: Sitting in a cold shower, ripping up $100 bills.

The fact they can get something with a sail to operate in anything larger than a bathtub for $4900 has me impressed!

Re:how about something a bit simpler

[jacquesm]

you're assuming quite a bit from what little I wrote.

for one I don't spend much time on my ass, rather the opposite, I wished there was more time in a day so I could get more work done. This post was actually made between cutting a bunch of grooves in brickwork, the dust needs to settle...

Second, a house building robot is one of my pet ideas and I've been mulling over the more complicated parts (raw materials delivery, on the spot mixing and curing of concrete, avoiding the fouling of the mechanisms and such).

Also I doubt I could do the job described above better (and I did not claim that I could), I'd just like to see less 'cool' stuff done with robotics. Vehicle control is where the sexy stuff is (camera input, navigation) and it's also where the funding is (see the various darpa challenges), but the benefits of something that can provide durable housing at extremely little cost are much more concrete (pun not intended) than the ability to *sail* a boat from one end of the planet to another by robot control.

Some of the more interesting ideas I've seen involved fusing sand using solar power, the energies involved are pretty scary though. The nice thing is that in those parts of the world where this kind of tech is needed most solar power and sand are pretty abundant.

Re:how about something a bit simpler

[ColdWetDog]

Have you ever been involved in a) robotics or b) house-building? Combining robots' limited capacity for detecting and dealing with "messy" and unpredictable situations with the reality of putting things together outside of pristine, well-supervised factory conditions is not going to work out all that well, at least with the current levels of technology.

What runs through my mind here is a new form of battle bot -

General_Contractor_bot: "Where the hell is the Framing_bot?"
Electrician_bot: BEEP! "Don't know, go find him yourself you mindless droid!"
General_Contractor_bot: "Call me a droid, will you?!"

(Sparks fly)
Meanwhile Plumbing_bot runs in circles because it talks only in ObjectiveC and doesn't understand anyone else. Besides, it doesn't want to get dirty.

I think this sort of thing would be the most likely scenario..

Re:how about something a bit simpler

[Ulven]

90 foot?! That's huge! 20 odd years ago Dad sailed the Atlantic several times in a 36 foot boat, and most of the other boats he saw were smaller.

These days, 36 foot is pretty much near the smaller end of the scale, but 90 foot is still awfully big to be considered the minimum size for the Atlantic.

Re:how about something a bit simpler

[Ulven]

If you're talking about Beluga Skysails, they can definitely operate under a wider range of conditions than you say.

The kites operate at anything up to 50 degrees to the wind, and are controlled by computer.

Re:how about something a bit simpler

[F�an�ro]

But if you are using the robots as blocks for your house then you can only use them for one house at a time, and building an ordinary house is several orders cheaper than building those robot bricks.

I do not see that change any time soon - In the foreseeabe future building a house out of self-assembling robots will continue to be much much more expensive and resource-consuming than the regular way.

so, while those self-assembling robots may be more interesting (and much more complicated) than atlantic-crossing robots, both are pretty much equally usefull for building houses for people to live in.

Re:how about something a bit simpler

[CastrTroy]

I don't really see the problem with it, but I don't really see why you would want to either. For the reasons I listed above. Sure you could build a house building robot, although it probably wouldn't do the entire job autonomously. Roof trusses already come preassembled in most cases. I don't see why you couldn't have robots assembling them in the first place.

Re:how about something a bit simpler

[Anonymous Coward]

I'd imagine navigating can vary between monotonous and dull to requiring some quick wits to avoid obstacles. The obstacle avoidance part would be more engaging, but I could see how long, drab voyages could lead a man to drink, crash into something and spill a lot of crude oil or whatever cargo is on board.

House building is always active, and requires hiring a lot of people. The option of augmenting or replacing some people who might get too bored to properly steer a craft worth millions with a cargo of billions (I just made up those figures, I don't really want to research those things =). Hiring laborers and contractors to build structures keeps economies afloat.

And if you need housing in a pinch, robots building things seems a lot more work than some easy-to-assemble kit that could be dropped in a remote, war-torn, or nature-ravaged area than dumping off a load of robots that you'd later need to retrieve.

Re:how about something a bit simpler

[RKThoadan]

Actually, building houses in relatively pristine factories and assembling them on-site (modular housing) is becoming quite popular, even for higher end houses.

Link to book "Modular Mansions" on Amazon:
http://www.amazon.com/Modular-Mansions-Sheri-Koones/dp/1586857126

Re:how about something a bit simpler

[Em Adespoton]

On the other hand, this would make remodeling MUCH easier :) Just hope your house doesn't get a virus while you're taking a shower.

Re:how about something a bit simpler

[Em Adespoton]

Actually, some houses are already framed by robots; there are already robotic log house framers -- robots shape logs to precise sizes, robots examine logs for stress points and faults. Other robots make join cuts in logs and robotic cranes move it all into place. Of course, they then generally take the house apart again and a framing crew reconstructs it on-site. But robots are already pretty heavily involved in many aspects of house building.

Re:how about something a bit simpler

[DrVomact]

Robotics challenges are usually somehow tied to military objectives such as navigating a certain terrain, rarely do they focus on something constructive and creative.

So navigation is a military technology? Of course warships have to be capable of navigation, but that's kind of like saying that because infantrymen must walk long distances, hiking is a form of military training. Nowhere in the article does it say this is a DARPA project...but even if they were kicking in some money, remember that your ability to post your comments is based on a "military" research project. If you don't remember ARPANET, Google it.

Of course, we should be wary—the favored contestant appears to be a product of that infamous think-tank of military inventiveness, Aberystwyth University in Wales! And you think devising a robot that knows how to navigate a sailboat across an ocean is trivial, and uncreative? Obviously, you've never sailed a boat across a pond.

Personally, the first thing I thought of when I saw the headline was that maybe we're getting close to the day when huge robotic sailing ships transfer cargo across the sea lanes, instead of those nasty diesel-burning freighters we have today. I think that would be kinda cool...and useful, even.

Re:how about something a bit simpler

[wfstanle]

Crossing the Atlantic requires going in a straight line for as long as possible.

Do you really know what you are talking about? Sure tacking into the wind requires a zigzag course, but with GPS navigation systems and good navigation software, this shouldn't be too hard. Even variables such as changes in wind direction and wind strength can be taken into account. As far as obstructions (even moving ones such as ships) can be handled with a radar system.

Re:how about something a bit simpler

[kabocox]

Like a robot that builds a house or so. A bit more useful too...
Robotics challenges are usually somehow tied to military objectives such as navigating a certain terrain, rarely do they focus on something constructive and creative.

Humans are much, much cheaper and easier to program than robots. You can hire illegal immigrants that can barely speak your language, yet they'd know enough to ask the going rate to mow your yard/clean your house or whatever and can find your home on a set schedule. I'd love a robotic lawn mower or maid, but they just are not as flexible as the stupidest/cheapest humans are. Those robotic mowers are really expensive. I can pay a neighbor kid $15-20 to do it and they'd mostly do a good job.

Here is a thought for you. Why the heck do we still have humans on garbage trucks? Because that task is currently far too difficult for robots to do. The same thing applies to truckers that just drive from point A to point B and don't do the loading/unloading of their truck. Why isn't that task automated? Because we don't have AI smart/safe/cheap enough to drive a commercial truck on interstate and refuel at the correct spots.

Why do people still drive? Well, we've discovered that task is very difficult to automate.

Re:how about something a bit simpler

[moosesocks]

Do you really know what you are talking about? Yes, I do know what I'm talking about, and agree that these might not necessarily be huge difficulties to overcome.

I maintain my original argument that you can't sail across the atlantic in a straight line, which was all that I was stating in my original post.

If you want to get really advanced, choosing the "optimal" course to sail along might actually be a fairly interesting problem to solve computationally, if you want to take meteorological data and forecasts into account, and update them along the way to choose the best course, while also avoiding lulls, obstacles and storms.

Re:how about something a bit simpler

[Trackster]

Not necessarily so. For a single Bot to build a house in the traditional manner, it would be pretty tough from a mechanical design standpoint. For multiple Bots to build one in the traditional manner it would be pretty easy. Bots building cars has to be a lot tougher than building a house which requires a lot less precision.

Also, having Bots build a house using alternative methods like the method described here or others would make the task much, much easier.

Re:how about something a bit simpler

[Amouth]

it is the min size for me to feel safe in rough seas..

i know it can be done on much smaller boats.. but i wouldn't want to be the person doing it..

i have done my fair share of risky/stupid things.. time to take it easy for a while

Re:how about something a bit simpler

[zippthorne]

How is building a house in any way simpler than navigating a vast, mostly barren body of water?

Re:how about something a bit simpler

[mdozturk]

As somebody who has spent time on the helm of a sailboat sailing by myself: it is hard to sail at 100% efficiency, but pretty easy to sail at, say, 50%.

require extremely quick and accurate responses to avoid capsizing

If you make the keel big and the sails small capsizing is a non-issue. I'm sure the boat was designed for stability NOT speed.

Having an unmanned craft sail from Portugal to the Caribbean is more complicated than landing an unmanned craft on Mars.

Bullshit. All they are doing is enter a course into the raymarine autopilot telling it to follow wind direction (for example: keep wind always at beam). Periodically the "robot" does course correction and trim the sails.

Re:how about something a bit simpler

[Detritus]

Why not build the house off-site? I recently watched several documentaries on modern naval shipbuilding and I was impressed by how the ships were composed largely of modules that were built off-site and shipped to the main shipyard for assembly and integration.

Re:how about something a bit simpler

[jacquesm]

because you'd need to ship the house, shipping a robot is a one time affair, it should be able to build as many houses as possible until it fails.

Usually raw materials of some form or other are already present at the prospective building site.

Re:how about something a bit simpler

[CaseyB]

Smaller might be better in this case. It's not hard to build an unsinkable boat below a certain size. The worst hurricane in the world can't sink a rubber duck.

Re:how about something a bit simpler

[DarkOx]

This is very useful though. We live in a global economy where goods are shipped all over the planet. We also getting near the brink where the fule that drives that shipping oil is going to get really expensive.

We might have to go back to more local economies that are less specialized if we don't find a cheap way move stuff.

Well this could be it. Sailing is pretty cost effective energy wise, but its SLOW, maybe 10 knots for a larger cargo vessal. People don't want to be at sea for months crossing the Ocean out of sight of land. If some robots could sail a ship safely and reliably we might beable to do lots of our shipping that way. Goods could stay in the pipeline for a while, you just need a steady stream.

Re:how about something a bit simpler

[daviddennis]

Well, this is a robot, so it wouldn't feel the waves.

You can technically sail the ocean on a Pacific Seacraft Flicka 20. You probably won't enjoy the experience, but if you could substitute a robot for you, it should do just fine in such a boat.

One question I have about this is that I believe it has already been done. There are systems available now that integrate an autopilot with GPS and radar, and automatically sail a specific course until a specific position is reached, and then change course until the next position is reached and so on. There are also devices that automatically use the wind to set the boat on an optimal course based on getting the maximum speed within certain parameters (these are called windvanes). With so many of the pieces already together this doesn't sound like a project that would be all that difficult.

That maximum speed of 4 knots certainly doesn't make it seem worth it, though. Might be more practical to use a diesel.

Avoiding obstacles would be an interesting question, though. To be a true solution you would have to change course automatically if the system saw a freighter. A lot of single-handed cruising sailors take the risk of running autopilots overnight but they are playing roulette with their boats and lives.

The kind of computing related to finding the obstacles considering a boat crashing up and down with the waves might indeed be tough.

D

Re:how about something a bit simpler

[Anonymous Coward]

deep sea vessels use meteorological data and recommended courses from weather routing services already, it saves them quite a bit in fuel costs.
ie; http://www.spos.nl/
and i agree with you about sailing in a straight line as the shortest distance between to places is actually a "great circle" (unless you're heading due north or south) but it's pretty simple to work out.
But somehow I doubt that the robots would be getting weather routing data, it would make it too easy, they should be able to make fairly simple (not particularly accurate) weather predictions using barometric pressure and wind speed/direction.
I hope they are well lit and have a good radar signature, i don't want their rigging wrapped around my propeller.

Re:how about something a bit simpler

[MichaelSmith]

In the foreseeabe future building a house out of self-assembling robots will continue to be much much more expensive and resource-consuming than the regular way. The cost of labour is already a significant part of the price of a new house, and labour rates are rising everywhere in the developed world.

Re:how about something a bit simpler

[MichaelSmith]

On the other hand, this would make remodeling MUCH easier :) Just hope your house doesn't get a virus while you're taking a shower. I have this vision of intelligent stacks of building materials ambushing homeless people in the middle of the night and building a home around them.

Re:how about something a bit simpler

[F�an�ro]

sure, but there are loads of other options: for example house-building robots that use cheap pre-fab components to build a house.

heck, even robots that build the whole house out of brick and mortar should be far easier to build and more economical than self-assembling robot bricks.

My point is that self-assembly is not relevant for house-building, robots may still be useful

Re:how about something a bit simpler

[KlausBreuer]

>Definition of sailing: Sitting in a cold shower, ripping up $100 bills.

If you're willing to play races in modern plastic (GFK) boats, yes, you're right.
However, very many people have lived beautiful lives on boats. Poor people, with little money. Lovely steel long-keeled beauties of sailing ships.

I grew up on one in the atlantic, and we didn't exactly swim in money ;)

Designing a sailing robot is a great idea, as sailing can be quite a bit more complex than you seem to think... and the little bots could be used as autopilots later. Nice to get some sleep if you're sailing alone...

Re:how about something a bit simpler

[Corwn of Amber]

How comes Roland still writes for /.? Is it not evident enough he's fried his brains on drugs? Did he ever post any article totally rooted in reality? His "science" is on the level of the justifications for supernatural stuff in X-Files. As in, "it does Not Work Like That in the Real World".

I'm going to learn whatever language Slash is written in just to tag all further Roland's stories with "ohnoitsroland", because that oh-so-helpful tag is so often forgotten.

The Pinta???

[AndGodSed]

Sleep now in the fire!

Natives are done for

[explosivejared]

Hmmm... historic trans-Atlantic journey by sea. Seems history is repeating itself.

If the white men hadn't done enough to the natives already... well then the coming robotic horde will mop up the rest. To all my indigenious friends out there, they say they come in peace now, but remember the last time you heard that.

Re:Natives are done for

[njcoder]

What makes you think they're not coming for us?

Re:Natives are done for

[Anonymous Coward]

What, you mean there are actually still some natives left in America?

Re:Natives are done for

[jellomizer]

I for one welcome our new Robotic overlords.

Re:Natives are done for

[Anonymous Coward]

This takes us one step closer to robot pirates. Pit them against ninja zombie gorillas and you'll have all the ingredients for a nerdgasm felt around the world.

Re:Natives are done for

[djones101]

Wouldn't it be...

I, for one, welcome our new robotic sailing overlords.

Re:Natives are done for

[Em Adespoton]

Just remember to tell their GPS units they're in India; that way it might take them a while to realize there's an entire continent to the north ;)

Not as difficult as the DARPA Grand Nationals?

[morgan_greywolf]

Is it just me or does this seem less difficult than the DARPA Grand Nationals. Disclaimer: I know next to nothing about sailing, despite the fact that I live near the ocean.

Re:Not as difficult as the DARPA Grand Nationals?

[Ngarrang]

Going in a straight line will be easy with the wind on your tail, but...Tacking seems to require more than just logic, but an intuitive sense of timing, direction and guess work. I think it will be interesting to see how the computer systems handle that aspect over the long-haul.

Re:Not as difficult as the DARPA Grand Nationals?

[Anonymous Coward]

Except the DARAP grand challenge lasted a day for the desert crossing, all the robots had to do was last that long. Here they have to last months without anyone attending to them.

Re:Not as difficult as the DARPA Grand Nationals?

[morgan_greywolf]

Except the DARAP grand challenge lasted a day for the desert crossing, all the robots had to do was last that long. Here they have to last months without anyone attending to them.

Very good point. I hadn't thought of that angle. Hopefully none of the entrants will be running Windows. *ducks*

Re:Not as difficult as the DARPA Grand Nationals?

[vertinox]

I know next to nothing about sailing, despite the fact that I live near the ocean.

I hardly know much about sailing other than history interest in sail travel during the 1500's through the 1700's and from my understanding sail travel is quite difficult compared to your standard propeller travel.

Especially if the wind is blowing the the opposite direction you want to go. One of the key inventions that did allow travel between Europe and the new world was the triangular sail which mitigates the issue by allowing the ship to travel at an angle sort of towards the destination even though the wind is blowing in the wrong direction. This is why the compass, sextant, and various other navigational tools were so important back then because since you couldn't often travel to your destination as the crow flies in a straight line you had to often find out where you were and then turn at an angle to aim towards your destination.

Of course if the winds were favorable, you could just plop up some square sails and enjoy the ride.

The key for these robots would be to able tell the wind direction and adjust accordingly which is a good deal easier with with GPS these days.

Re:Not as difficult as the DARPA Grand Nationals?

[Ulven]

Just to be nitpicky, but square sails could sail into the wind, just not very well. I think they managed about 60 degrees off the wind. Bermudan or other triangular rigs can get closer to 45 or 40 degrees.

Re:Not as difficult as the DARPA Grand Nationals?

[zippthorne]

It is not the triangular sail (fore and aft rigging, really, regardless of sail shape) which allows you to travel up wind, but the keel.

Tacking is actually two-fluid sailing, which implies that you need a sail in both fluids (and, obviously, a velocity difference between them also). Of course, with the density of water, the wet-sail doesn't need to be nearly as large as the air-sail, and with small enough boats, the hull itself acts as a fairly inefficient keel.

You do need to be able to rotate the sails, but square-rigged vessels are perfectly capable of this, albeit not necessarily to the degree that a Bermuda-rigged sloop would be able to.

I say old boy...

[Linker3000]

"The Times of London"? Never heard of it.

You mean "The UK national paper 'The Times'".

There's more to the UK than Buckingham Palace, tea at the Savoy, Harrods and and Big Ben, Mr P.

Re:I say old boy...

[morgan_greywolf]

There's more to the UK than Buckingham Palace, tea at the Savoy, Harrods and and Big Ben,

Like say, the London Bridge? Oh yeah, isn't that in the middle of the Arizona desert now?

Re:I say old boy...

[Linker3000]

Indeed old chap - we let you have the old one.

Mind you, I hear rumour that the poor old buyer mistakenly thought he was getting our dear Tower Bridge.

Silly Sausage.

Re:I say old boy...

[catbertscousin]

The trouble with famous landmarks is people tend to think they're the only thing like them in that area. When my brother was showing his pictures from his senior trip to London, he had several long discussions with people trying to explain to them that there was, in fact, more than one bridge in London. Shock and disbelief.
I've also met several people who thought New York was just the city; couldn't convince them that Albany, Buffalo, and Rochester were also New York. *sigh*

And

[William Robinson]

"Keep sailing. After 3 months we also have chance to be called 'Hello, sailor' "

It's a bit small!

[Chief Wongoller]

This boat is only 3.65 meters long - that's a mere twelve feet, which is smaller than many dingies I have sailed. Normally sailing craft have to be much bigger to withstand the ferocity of ocean winds and waves,which simply swamp craft of this size. So how can it possibly stay afloat for several months?

Re:It's a bit small!

[MichaelSmith]

This boat is only 3.65 meters long - that's a mere twelve feet, which is smaller than many dingies I have sailed. Normally sailing craft have to be much bigger to withstand the ferocity of ocean winds and waves,which simply swamp craft of this size. So how can it possibly stay afloat for several months? Exactly like a submarine (or a shipping container).

Re:It's a bit small!

[Lonedar]

Well, I would imagine that the lack of crew lets the designers make the boat more robust - e.g. you can keep the hull completely watertight, as you don't need a cabin, portholes, hatches etc. Also, if they design the boat with a heavy enough keel I would imagine that it would be resistant to capsizing.

Re:It's a bit small!

[ZorbaTHut]

If it's buoyant, watertight, and has an appropriate center of gravity, then it'll usually right itself if it capsizes. If it's equipped with some device to "flip it over" on the off chance that it doesn't do so automatically, it could easily make it the entire way - the only risk would be damage from storms or running out of power.

Re:It's a bit small!

[moosesocks]

Not necessarily if it's a sailboat.

A boat that size usually depends upon the weight of its crew to keep it balanced. Similarly, unless it's got an absolutely immense keel, it can easily tip over into the water.

In the event that the boat completely inverts itself, which is fairly likely because the weight of the sails and mast often account for a considerable portion of the weight of the craft, it could become virtually impossible for the boat to "right itself". Also remember that the sails generate a good deal of underwater "resistance" that make it even more difficult to right the boat.

Re:It's a bit small!

[harry666t]

Maybe it's balanced properly and made unsinkable & completely waterproof?

A few years ago, when I was more into sailing, I thought of a RC model of a yacht with all the stuff like setting & adjusting sails, balancing, steering, done via small RC engines... Well, it was just a thought, but all seemed reasonable and doable.

Re:It's a bit small!

[UnderCoverPenguin]

With no human (or, presumably other animal) crew, it needs only a sealed capsule for the electronics, batteries, etc, and for flotation. As long as the boat can stay more or less upright - or can right itself - it will be able to continue sailing.

Re:It's a bit small!

[Anonymous Coward]

Quit bragging about the size of your dingies.

Re:It's a bit small!

[david.given]

Not necessarily if it's a sailboat. A boat that size usually depends upon the weight of its crew to keep it balanced. Similarly, unless it's got an absolutely immense keel, it can easily tip over into the water.

Actually, building self-righting / uncapsizable boats is pretty straightforward. Remember that the keel needs to be heavy enough to offset the tipping moment of the sails; normally this means they're really, really heavy. Also remember that the keel is submerged in water, which means that its effective weight is rather lower than it would be in air.

With a bit of forethought, you end up with a boat which will tip over until the keel starts coming out of the water, and then it'll just stop --- any additional heel will cause more keel to emerge, which will cause the effective weight of the keel to increase hugely, which will prevent any further heeling.

Even if by some miracle you do end up with the boat upside down, it's unstable in that attitude and will right itself. Yes, the sails will cause huge water resistance, but that resistance is proportional to the speed of motion through the water; it won't stop the self-righting, it'll just cause it to happen slowly. (Also, the sails will act to prevent the capsize in the first place, for exactly the same reason.)

What tends to happen these days on decently designed boats is knock-down; a gust of wind causes the boat to be knocked onto its side, up to the point where the keel's righting moment offsets the tipping moment of the wind against the sail. This can be very hazardous to the crew, but hey, no crew! When the gust passes, the boat will right itself (usually even if it's filled with water).

The biggest risk is that all this process is extremely violent; the boat's being slammed about hugely. You run a very real risk of bits of the boat actually breaking. The tension at the base of the mast is huge at the best of times, and if the mast breaks under strain and doesn't come completely free of the boat it can very easily smash through the bottom of the hull. Which Would Be Bad. That's one of the reasons why people like unstayed masts these days; if you get dismasted, you don't end up with a huge, heavy, sodden and very dangerous lump of stuff smashing about on top of your boat --- you're much more likely to lose it completely overboard. Much safer.

While this does tend to apply to yachts rather than dinghies, which as you say largely use humans for ballast, you really do get yachts that size --- the difference is largely design rather than size. My father designed, built and sailed a highly successful yacht only a little bigger --- 15 feet, I believe. It was a bilge keel gaff rig with two monster lumps of concrete for the keel, and slept three. It would heel comfortably to about 45 degrees and then just stop. My father tried quite hard on several occasions to get the cabin windows in the water (much to my horror) and failed every time...

Re:It's a bit small!

[Hognoxious]

A boat that size usually depends upon the weight of its crew to keep it balanced.

The ones that are designed to carry a crew do.

Re:It's a bit small!

[Bloke down the pub]

This boat is only 3.65 meters long - that's a mere twelve feet

In America they have bigger canoes - and that's just the width!

Re:It's a bit small!

[CmdrGravy]

I'd imagine so long as it had enough buoyancy it could stay afloat indefinitely so I imagine the real challenge would be to generate enough power to maintain a course and not just drift. Of course the mast could break or the sail rip or whatever I suppose which might put to an end to things.

Re:It's a bit small!

[KokorHekkus]

Actually, if it's smaller it can propably withstand the ocean forces more easily in most cases since there will be less chance of the forces finding something in the construction that will provide leverage. Just take a pencil hold it with your fingertips at the end and snap it off, it should be pretty easy for most people. Then try doing the same thing to an inch long pencil stump.

And with a smaller boat you can easily build an almost unsinkable craft if you use a sandwich-type hull filled with enough floatation material so that even if the hull is completly waterfilled the boat will not sink. This was what Sven Yrvind used in some of his constructions ( http://en.wikipedia.org/wiki/Sven_Yrvind )

Re:It's a bit small!

[Anonymous Coward]

I bet they'll use batteries instead of concrete this time :)

Re:It's a bit small!

[ChrisA90278]

Yes it's a small boat but it does not have to cary humans so it does not ned things like companionway hatches, food and water. The boat I'd image would be completly sealed and heavy blasted with lead acid batteries. I imagine the boats will be self righting. If I were designing them they's have rigid sails, more like an airplane wing than a sail. A boat like that simply coud not turn upside down

Re:It's a bit small!

[Anonymous Coward]

What part of and has an appropriate center of gravity don't you seem to understand?

Re:It's a bit small!

[Anonymous Coward]

Hey, nice post. I'll add just a little.

Is the sail cloth, or a rigid wing? I scanned the article quickly, and I did not see that mentioned.

With a rigid wing, the pivot axis can be attached to the drive mechanics through a clutch with preset drag. In the case of a gust, the clutch will slip, lowering the load on the pivot axis. This lessens stress on the hull as well as reduces the chance of tipping the craft. The rotary encoders are affixed to the sail axis, not the drive. So the drive tries to return the sail to the preferred position after the gust displaces it about the axis of rotation. The feedback mechanism is pretty simple, and only consumes power to recover after an anomalous event. You could even encode for 'gust events' and stop sail movement until those event counts / time drop below a threshold. That would lessen stress and reduce power consumption.

I see this as a great design challenge. With the price of fuel and the desire to drop CO2 production, wouldn't it be nice if we could have mid-sized pleasure craft that could be single handed through applied robotics? As an old sail hand that has spent years in 14' and 16' craft, I think a single handed 40' cruiser would be great! Particularly if it could remain underway through the night while the captain slept.

I, for one, do welcome our robotic sailing overlords.

Re:It's a bit small!

[moosesocks]

Mod parent up. He makes a few very good points.

My main gripe still is, though, that these boats don't look quite big or heavy enough to have a keel big enough to do what you're describing.

You're also absolutely right about bits of the boat breaking under stress. A few months ago, I saw a big gust break one of the stays off of a Firefly, subsequently causing the mast to snap in half, under what were otherwise fairly reasonable conditions. This is particularly notable, considering that Fireflies have a fairly small sail area, and a big beefy aluminium mast.

"creative destruction"?

[Apatharch]

That's almost as bad an oxymoron as 'military intelligence'.

Re:"creative destruction"?

[Anonymous Coward]

In what way are "creative" and "destruction" in any way contradictory?

Whalesong?

[Goffee71]

Hopefully the robo-boat will be sung at by whales, learn their language and spread a message of peace and hope for mankind, while sending a signal into space for the whale's ancestors to pick them up. At which point the military will step in and blow it to bits. Now here's Larry with the Sport

Call me pedantic if you like, but...

[Apatharch]

According to the Times article there are actually eight robotic craft competing - the Pinta and seven others.

Got to go to a tropical island for three months

[MosesJones]

Now that is genius. Aber for anyone who doesn't know is one of the coldest, wettest, windiest and bleakest places in the UK, its okay in the summer but these students and their prof have just come up with a reason to be on a tropical island for three months "you never know when it might actually arrive".

Cheap booze, great weather, women in bikinis and no threats from the druids... brilliant.

Re:Got to go to a tropical island for three months

[mgblst]

...Aber for anyone who doesn't know is one of the coldest, wettest, windiest and bleakest places in the UK

Really, I though that there were only two places in the UK like that, Scotland and England. Maybe Wales as well. And most of the time Ireland.

Oh well, i should be happy, we had one week of summer this year, a record I hear!

Back to the story, this is a great idea. If they find someone to power themselves, you have loads of drones in the ocean, monitoring ocean currents, as well as a border patrol (great for places like Australia).

Re:Got to go to a tropical island for three months

[GingerDog]

well, at the very least they have to go to France for the launch anyway :)

The professor involved did spend most of a year sailing around the Carribean a few years ago, so you might not be far from the truth there :).

I've heard it's undergone some considerable testing, so would hope it'll at least get half way (a bit like Beagle the space ship thing)

Re:Got to go to a tropical island for three months

[Anonymous Coward]

You get plenty of cheap booze in Aber. Triple vodkas in Lord Beechings for a quid.

And it's not so cold, being on the coast. It gets a bit of rain, but I wouldn't describe it as "bleak".

Beagle B robotic boat

[OzPeter]

TFA mentions a larger robotic boat called the "Beagle B". This name sounds suspiciously like someone is paying homage to the Malcolm Jameson story "Children of the Betsy-B" that was first published in Astounding Science Fiction in 1939.

Re:Beagle B robotic boat

[Goffee71]

Since the last famous Beagle vessel smacked into Mars at high speed, I wouldn't put any money on this one... hopes Colin Pillinger doesn't lurk on /.

Re:Beagle B robotic boat

[Eustace Tilley]

Charles Darwin's second voyage was on the HMS Beagle.

Re:Beagle B robotic boat

[Anonymous Coward]

Obviously (maybe not), you are supposed to think that this is a reference to the Darwin voyage, but in fact this is the boat built to transport the robotic telephone sanitizers and the marketing droids, who insisted on having a go - it is a bigger boat so that the oil bath can be accomodated.

Call me even more pedantic if you like, but...

[pieleric]

Its designers hope the Pinta will become the first robot to cross an ocean using only wind power. But from the article:

The boat uses solar panels to provide the power to operate[..] So I'm afraid it will not be only wind powered.

Re:Call me even more pedantic if you like, but...

[anno1602]

To take the pedantry to its logical conlusion: Propulsion itself will be by wind power, but the power adjust the sail(s?) and for the computer will come from the solar panels.

at maximum speed?

[dwater]

> Pinta is expected to sail for three months at a maximum speed of four knots (about 7.4 kph).

Why always at maximum speed? Does the switch only have two positions - 0, and maximum?

The size of the boat may be a hinderance...

[misterthirsty]

but it should not adversely affect the ability of these boats to do a trans Atlantic crossing. Modern designs of sailboats are self righting, and there are several historical examples of small boats crossing large area's of water. Lt. Bligh of the Mutiny on the Bounty fame sailed about 3600 nautical miles in an open boat, the Polynesian Islanders have been doing this for centuries, and some guy recently crossed the Atlantic in a boat the size of a bath tub. Here is a pretty good list of small boats going long distances in open ocean.

As for these boats being robotic, I think it is a great test of ingenuity, combining modern technology with the oldest and most time tested form of long distance travel. I would only worry about one of these things being run down by a freighter or similarly large vessel, as a human presence on board is the best way to avoid collisions at sea...

Re:The size of the boat may be a hinderance...

[fatcop]

I'd place more money on the first boatload of passers-by nabbing it as a souvenir or an Xmas present for their kid. Tanks very mooch.

Three words

[Anonymous Coward]

robotic boat safari

Deer slugs ought to do it.

Message in a bottle (MIB)

[aldm]

Can they spill some MIBs for the old times sake?

scooner

[Is0m0rph]

Look a scooner! You dumb bastard that's not a scooner it's a sailboat.

Re:scooner

[Anonymous Coward]

Awesome freaking movie! :-D

You face forward, or you face the possibility of shock and damage. LOL

Durgs and stuff....

[slptec]

I smell a new way to transportdurgs, nukes and other nice thingys!

What about liability?

[Anonymous Coward]

Having looked into this a few years ago, there's a real issue with Autonomous watercraft, in terms of their legal treatment. What happens if a supertanker or cruise ship happens to have an accident and your little widget is in the area? Just how much liability insurance are you carrying? The law of the sea doesn't appear to accommodate autonomous stuff very well.

Good Luck!

[MrJerryNormandinSir]

This is cool! I've built smaller robots that utilize an digital compass and dead reckoning. This robotic sailboat however just blows my mind. I know a little about sailing, I'll be very interested in the algorithm they wrote to control the sails, create adaptive waypoints to make up for wind changes. After all, this is a sailboat. I'd be interested to see how well the bot tacks! Collision avoidance, I'd be interested to see how the radar hits are processed. This is an amazing undertaking.
Good Luck!!

Navigational errors

[Anonymous Coward]

They had to be reprogrammed on the fly, because they kept thinking Massachusetts was India.

prepareToBoardArrrr()

[Doc Ruby]

How long before the first robosail is snatched by human pirates on the high seas? An unmanned robot boat will have to have a lot more extensive (and expensive) AI and "interaction" HW than a defenseless, naive one.

And how long before the pirates relaunch some of those captured robot boats back at us, with that interaction HW designed as a new defacto industry standard, regardless of any ISO specs?

Re:prepareToBoardArrrr()

[rholland356]

While pirates might take an interest, I should think this technology will prove very useful to the drug traffickers of the world.

Re:prepareToBoardArrrr()

[Doc Ruby]

If the robosails (or motorboats) can dump the contraband, especially with some kind of self-destruct, then smugglers around the world will surely appreciate it. I can see the current generation of Fatherland^WHomeland Security marketers making a case for completely "closing" (with invasive surveillance and expensive interdiction) all maritime boundaries, and even monitoring boats (live and robot) that go outside the boundaries to where they could meet a roboboat in international waters.

In fact this "SF movie threat" could be an easy argument for total global surveillance, an excuse for tracking literally everything that moves on 72% of the Earth's surface.

"Funny" how the same people who will make that argument will also make the argument for completely unchecked international trade. And already refuse to fund securing ports, like here in NYC where we've got the country's #1 or #2 largest port (depending on measurement unit or quarterly business) and the #1 proven terrorist target and longtime smuggler's crossroads.

Collision Avoidance

[Anonymous Coward]

The rules for the trans Atlantic race state, "Boats must take appropriate precautions to avoid collisions. This might include the use of radar reflectors, brightly coloured panels, navigation lights or warning labels/flags."

I'm a cruising sailor who does blue water sailing often. I've got to tell you that this idea spooks me.

It has always been possible to have vessels "not under command" In fact, the rules for navigation lighting have always provided specific ways to mark such vessels so that you know about it.

Single handed sailors obviously need to sleep sometimes, so at least part of the time they are not under command.

I also suspect, but I can't prove, that many ships at sea proceed illegally with nobody awake or nobody sober on the bridge. I've tried and tried to hail them on the VHF radio, and they don't respond. They are loose cannons that from time to time run over hapless smaller vessels without even noticing.

I guess that small robot sailboats aren't much of a real risk for cruisers. Still, the idea freaks me out. Suppose they are very successful, and in a few years, the oceans are filled with thousands of them?

Sailing

[PhotoGuy]

I love sailing. I find it to almost be an art, managing the interactions of the wind and the water to make a vehicle move, while watching for the best route (especially when racing), managing and training the crew, and enjoying the splendor all around you.

One of the aspects I love about sailing, is the challenge of dealing with dozens of inputs (wind direction, wind speed, boat heel, current, etc.) and controls. Most people don't realize the level of detail with which one can adjust a sail. While airplanes are stuck with a fixed aerofoil, sails can be adjusted by stretching the front (luff), the back (leach), the bottom (the foot). You control these three sides with the halliard (raises the sail), downhaul (pulls down on the sail, easier to tighten the luff after the sail has been raised), outhaul (tightens the foot), leech line (tightens the leech/back of the sail), boom vang (pulls down on the bottom of the sail). With these, you can set the depth and shape of the sail to accommodate the current wind. (Heavier winds work better with flatter sails, lighter winds, with a bit fuller sails.) And of course you have to keep the proper angle of the sail with the wind by using the mainsheet, traveller, vang.

It really is a thing of beauty to get a sail working properly; then you combine that with a foresail (jib) that helps the flow over the set of sails. (There are often bits of yarn, ticklers, that help you see the flow over the sails, and see if it's laminar or turbulent.)

All that being said, pretty much every one of these many factors could be measured, analyzed, and appropriately adjusted by a computer and associated sensing/control hardware. And in some ways, seeing a system manage all those factors so accurately and elegantly is a bit of art in itself.

And there very few dangerous situations (wind coming around behind to flip the sail over in a crash jibe) that the computer and sensors could spot and deal with before they become a problem.

The main thing the computer lacks is the ability to appreciate the water rushing by the hull, the seabirds, the seals, the beauty.

It is still a worthwhile endeavor. Plus, the technology from such projects could filter down into products for sailors, who might be unable or unwilling to deal with a lot of the details. A lot of cruising sailors would love to have their sails trimmed properly by a computer. More power to them. It's not for me, I want to tweak every bit of the boat myself, for the joy of it; but if someone (including myself at times) wants to kick back and relax, while still having the boat perform, sure, let the computers do some work.

Drugs

[Anonymous Coward]

robotic drug running?

In other news...

[Anonymous Coward]

... robot boat missing, presumed stolen.