Tuesday, March 4, 2014

The drone that will sail itself around the world

 Saildrone, an unmanned, fully autonomous ocean vessel powered by wind and solar energy, recently completed a record-breaking ocean voyage, crossing 2,100 miles of Pacific from San Francisco to Hawaii.
A small fleet of Saildrones will soon be patrolling the world's oceans; they'll track sharks, stand guard over protected areas and gather critical data to help stop climate change.

From Wired

An Orange Pixel flickers on the horizon, sandwiched between the inky azure of the mid-Pacific and the robin’s-egg pale of the Hawaiian sky.
Richard Jenkins is the first to see it—a sailing robot, which has been blowing our way for a month.
We’re in a small motorboat 7 miles out at sea, just north of Oahu’s windward shore. Dylan Owens gets the next good glimpse.
“I see the wing,” he exclaims, “and the tail!”

 Photograph by Jos Cocquyt

Jenkins and Owens are the engineering duo behind Saildrone, which in the words of their website is “a wind-powered autonomous surface vehicle.”
On October 1, the 19-foot craft was set loose in the San Francisco Bay with a simple command lodged in its electronic brain: Sail to Hawaii.
For 2,248 nautical miles the boat did the rest.

The path it chose happens to be identical to that of the annual Pacific Cup sailing race, and the fastest anyone has traversed this course is just over five days.
The single-handed-sailing record is eight and a half days.
As Jenkins and Owens look on, Saildrone is about to complete what might be called the first no-handed ocean sail: San Francisco to Hawaii in 34 days.
It’s not quick, but then again there is no one aboard to complain.
The journey has included a storm with gale-force winds followed by two weeks of doldrums.
During the tempest, Saildrone was reporting speeds of up to 16 miles per hour and angles as extreme as 75 degrees, meaning it was heeled over and surfing down the backside of breaking waves—waves with enough power to snap it in two had they caught the boat in the wrong position.
The doldrums were equally worrisome: With no one aboard to scrub the bottom, algae, seaweed, and barnacles might have overtaken Saildrone, transforming it into just another piece of flotsam.

As the vessel sails into sight, I see that it’s a streamliner—a narrow hull stabilized by two outriggers, one on each side. Its “sail” is a sail in name only; in reality it’s a 20-foot-high, solid carbon-fiber wing.
Extending from the back of this wing, halfway up the mast, is a tail—just like an airplane’s.
(“That’s a little trick that I stole from the Wright brothers,” Jenkins says.)
Above the waterline the boat is painted safety orange and emblazoned with the words OCEAN RESEARCH IN PROGRESS in all caps.
The hull is black with bottom paint, and near the bow is the name in a fancy serif: Honey Badger.


The Honey Badger is more than a sailboat and more than a robot, although it’s both of those things.
The Pacific crossing is really a test of a new type of sail that automatically keeps itself pointed into the wind, like a weather vane.
Adjusting a little tab on the back of the tail—a task handled by the Honey Badger’s autopilot—is enough to maintain the correct course and to angle the wing so it creates forward thrust.
There’s no need to employ ropes, winches, or even sailors.
The mechanism is so simple it might really be best regarded as a plug-and-play power source.
Like a windmill, it converts a ubiquitous natural resource into usable energy.
Autonomous sailing craft could be used to gather information on the vast, untraveled reaches of the world’s oceans .
Its potential goes far beyond record-setting jaunts to Hawaii.
One obvious application is to mount the wing on a fleet of sensor-laden drones and send them sailing into the world’s oceans, where they could report on their findings.
“I want to get the data we need to show that global warming is real,” Jenkins says.
To that end, they could monitor ocean acidification, a key barometer of climate change.
Drones could replace the world’s weather and tsunami buoys.
The waters around oil platforms could be sniffed 24/7 for the first signs of a spill.
Tagged sharks, whales, and other marine life could be followed and their locations patched into the international marine-traffic control system with a warning to stay away.
Protected borders, coastlines, islands, and environmentally sensitive marine areas could be patrolled by drones programmed to photograph any interloping ships.


But scientific and security uses are just the beginning.
The biggest impact might be to industry: oil and gas, gold mining, diamond mining, fisheries, shipping, military.
For all kinds of reasons—regulatory, exploratory, classified—those industries need to know what is happening in the vast, nearly invisible reaches of the world’s oceans.
Tallied up, everything that is taken from or moves over the seas constitutes more than $2.5 trillion a year in business activity—nearly 4 percent of the total world economy.
What’s more, Saildrone’s technology is so efficient it could potentially power vessels that today require a motor.
Jenkins has developed a scaled-up version of his wingsail to propel passenger ferries that ply the waters of the San Francisco Bay.
On windy days, the ferry motors would power down while the wings did most of the work.
By the time you read this, the ferry wing will be flying from a test sled backed by two government agencies, sailing back and forth along commute routes.
Jenkins is confident the test will prove that in only a few years, the cost of retrofitting the ferries will pay for itself in fuel savings.

As we draw closer, it becomes clear that the Honey Badger has made the journey unscathed, and the mood changes from worry to jubilation.
“She looks just like I left her!” Jenkins, the boat’s mop-headed designer, says with genuine surprise in his voice. Owens—who is responsible for Saildrone’s electronics—is similarly relieved.
“This makes it concrete,” he says.
“For the past month it’s just been an icon on a web page.”
On the way into the harbor with the Honey Badger in tow, the men share Budweisers and congratulations.
“I was hoping to find a castaway hugging the back or a tooth from a great white or at least some guano on the deck,” Jenkins says.
“But it’s Honey Badger,” he continues, his cherubic face twisting into the froggy expression that always proceeds a joke.
Owens chimes in for the oft-repeated punch line, a quip from the viral video that gave the craft its name.
Honey Badger don’t care,” and then, with feeling: “Honey Badger don’t give a shit!”


Learning to design a yacht

Jenkins, 37, grew up in a sailing town on the southern coast of England.
He made his first ocean crossing at 16, working as a deckhand on a small yacht sailing from Bermuda to Spain.
At 17 he was toiling as a draftsman at a boatyard near his hometown of Lymington when he spied an abandoned wreck tucked away in the back of the yard. It was an odd type of boat: a land yacht, with wheels instead of a keel, and a hard wing instead of a soft, floppy sail.
The owner of the yard explained that he had built it for a customer who wanted to break the world land-sailing speed record—which at that time stood at 98 mph—but it had never been finished. Jenkins, sensing an opportunity to find fame and fortune, asked if he could have the boat.
The yard owner not only agreed, he also pretended not to notice all the carbon fiber that started going missing around the shop as Jenkins poured himself into his repair work.
“I thought it would take only a year or two to get the record,” Jenkins says.
In fact it took him a little more than 10 years and nearly got him killed a half-dozen times.
He rebuilt the land yacht in six months, and his initial attempts at a new record, which took place on an active RAF airfield known for its strong crosswinds, were very promising.
The problem was money: He needed cash to keep improving the boat.
“I was just a poor student,” he says.
He spent a lot of time meeting potential donors—he even met Prince Philip—and the resulting publicity alerted the world to his plans.
But by the time Jenkins hit 98 mph in the craft he had christened the WindJet, an American team had raised the record to 116. To compete, he needed a new, faster boat.

For the second craft, the WindJet Mark 2, Jenkins designed a much bigger wing and added IndyCar slicks.
The traction came in handy at about 100 mph, when Jenkins spied a military cargo plane swooping toward his runway with its landing lights on and its wheels down.
The Mark 3 was built of steel and tested on the high-traction salt flats of Western Australia.
On its second run, the lead counterweight at the front of the wing tore away from its mount, smashed through the cockpit canopy, and came close to braining a helmetless Jenkins.
On the maiden voyage of the Mark 4, the windward outrigger actually started to take off like a plane, nearly causing a catastrophic capsize.
“It is like gambling: It’s double or quits the whole time,” Jenkins says.

After four failed land-yacht designs, the situation was dire: The dry season was over.
Zero resources.
Every credit card but one maxed out. Jenkins had a decision to make.
Wait in Australia for another year, until the next land-sailing season?
Or pack the land yacht into a container bound for America and try there?
He went to America, and in the spring of 2009, on a dry lake bed just south of Las Vegas, he piloted a wind-powered sail rocket called the Greenbird to 126 mph.


 On the morning of March 26th, on the dry Lake Ivanpah, The Ecotricity Greenbird driven by British engineer, Richard Jenkins smashed the world land speed record for wind powered vehicles. The Greenbird clocked 126.1 mph (202.9 km/h) , eclipsing the old, American held, record of 116 mph , set by Bob Schumacher in the Iron Duck in March 1999 at the same location.

Jenkins was officially the fastest sailor in the world. Satisfying, yes.
But after nearly a decade spent camping alone in the desert, he figured that what he had gotten from the experience was mostly adventure.
“I was completely unaware that the wing technology I had evolved would be useful for anything except breaking land-sailing records,” he says.

The wing that Jenkins created for that final, successful land yacht had a tail. It’s not an obvious design. After all, most boats have soft sails, not hard wings.
And even the sailboats that do have hard wings—like the $10 million wingsailed racing catamarans that dueled in last year’s America’s Cup—don’t have tails.
The tail was the breakthrough idea that got Jenkins in the record books, it’s what got Saildrone to Hawaii, and it’s what has the potential to disrupt a multitrillion-dollar slice of the global GDP.
But to understand the genius of Jenkins’ tail, it’s necessary to go back to first principles.

Airplanes have tails.
And tails have horizontal-flight-control surfaces called elevators.
They govern the plane’s pitch and thus the angle of the wings relative to the air moving across them. This angle determines the lift the wings create.
More lift means the plane ascends; less lift, it descends.
Once Jenkins started reaching airplanelike speeds in the Greenbird, he realized he needed a machine that worked less like a sailboat and more like an airplane.
Sailboats have sails.
Aerodynamically speaking, a sail is a wing.
But the angle of a sail relative to the air moving across it—the wind, in other words—is controlled and adjusted by means of ropes and pulleys.
Tremendous force (and usually a winch) is needed to set a sail so that it cuts through the wind at the correct angle and creates the lift that moves the boat.
And then the problem becomes keeping the sail at the correct angle.
The boat may turn, which turns the sail with it.
The boat may speed up, which changes the speed and direction of the wind passing over it.
Or the wind may shift, changing speed and direction all on its own.
In every case, the angle must be readjusted manually—that’s what is meant by trimming a sail.

Once Jenkins started reaching airplanelike speeds in the Greenbird, he realized he needed a machine that worked less like a sailboat and more like an airplane.
At 126 mph, even the steadiest wind changes constantly as you blast across it.
No human sailor has reflexes fast enough to keep up with wind that shifts second by second.
The fastest racing sailboats in the world use wings—but they are still operated like sails, using ropes and pulleys and winches.
Greenbird’s wing works totally differently—it’s controlled by its tail, like the wing on a plane.
The Greenbird sails like an airplane flies, except that while the elevators on a plane’s tail send it climbing up or gliding down, the tab on the Greenbird’s tail makes its wing pull left or right.
It’s the same basic action, just rotated 90 degrees.

A hard wing on a free-rotating mount is a much more difficult thing to engineer than a mast—a simple pole held up by guy wires—but the payoff is in the actual sailing.
By severing all the ropes that run between the boat and the sail on a normal yacht, a lot of the complexity of sailing goes away.
In a normal sailboat, every turn of the rudder turns the sail.
Not so with a free-rotating wing, which by its very nature is always correctly angled into the wind.
Furthermore, dialing in the amount of sideways lift generated by the wing—thrust, in other words—is a matter of adjusting the elevator-like tab on the back of the tail.
The Greenbird had only two controls: the steering wheel and what was, in effect, a throttle.

How Saildrone Works

The 6 technology secrets that float this autonomous, ocean-­crossing boat. —A.F.
1. The Wing
As wind passes over it, the wing produces thrust. That force is concentrated on its axis of rotation, preventing the wing from spinning wildly.
2. The Tail
A ­little tab at the back of the tail can be set to the left or right, causing the wing to rotate a few degrees and maintain an efficient angle of attack.
3. The The Counterweight
Positioned at the end of a spar, it adjusts the wing’s equilibrium so its center of gravity is balanced, allowing it to rotate as needed.
4. The Rudder
While in theory it’s pos­sible to operate Saildrone by using only the sail, it’s more efficient to use a ­rudder to point the boat where you want it to go.
5. The Autopilot
GPS provides speed data and location. That’s all Saildrone needs to know. Navigation instructions reach the autopilot via satellite.
6. The Keel
If Saildrone gets knocked over, it will right itself because of the keel’s weighting. Its steep angle sheds debris like kelp and lost fishing nets.
(image : Brian Christie Design)

By the time he captured the land-sailing record, Jenkins had racked up a pile of debt.
To start working it off, he accepted an offer to move to San Francisco and help design a kite boat for Google cofounders Larry Page and Sergey Brin.
It was interesting work, and he was introduced to a whole new scene, one where out-there engineering projects like the self-driving car and augmented-reality glasses were the stuff of everyday life.
Suddenly his 126-mph accomplishment seemed pretty insignificant.
“No one cares about land sailing,” Jenkins thought, “but the first drone around the world?”

Jenkins realized that the wing he’d evolved for the Greenbird would be perfect on an oceangoing drone.
Its tail simplified the process of sailing so much that even a robot could handle it.
The bot would need only three moving parts: the elevator-like tab on the tail, the rudder, and the free-rotating wing itself.
What’s more, only two of those parts—the tail tab and the rudder—would need power.
A few off-the-shelf solar panels would provide more than enough.
Jenkins knew from long experience that the fewer parts there were, the fewer parts there were to break.
His oceangoing drone needed to be single-minded, bulletproof, and absolutely spartan.

It happened that Owens was working in the same boatyard as Jenkins was, though on a different project: writing software and building watertight, salt-resistant, pressure-tested controllers for a submarine project.
He too was daydreaming about building the first around-the-world drone.
Jenkins, a boatbuilder, met Owens, a hacker, on the shop floor in April 2010, and they quickly realized the obvious: They should quit their jobs and join forces.

David VS Goliath

The two didn’t actually form a company until they got a kick in the ass: News broke that a startup called Liquid Robotics had already launched an oceangoing drone.
The company had developed a line of boats it called Wave Gliders, which harvest their energy from the up-and-down movement of waves in the open sea.
An armada of four Wave Gliders set off from San Francisco in November 2011 to cross the Pacific for Australia.
The goal: Make it into Guinness World Records for the longest journey by an unmanned autonomous surface vehicle.
It was a publicity stunt—Liquid Robotics, which at the time was riding high on $36.5 million in VC funding, wanted some press. It billed its invention as “the wheel for the ocean.”
At first Owens thought that the wave-energy system was a pretty neat technology.
But the more he learned about it, the less sense it made to him.
Wave Glider’s top speed was incredibly slow: 2 mph in the best conditions.
Too slow to get out of the way of bad weather, too slow even to fight an ocean current like the Kuroshio or the Gulf Stream. And the wave-energy harvesting mechanism that hung beneath?
It was full of moving parts just waiting to get fouled by slime, kelp, barnacles, and wayward fishing nets.

Jenkins was similarly perplexed, not just because Wave Glider was the stupidest thing he’d ever seen, but because he couldn’t believe that anyone would pay $36.5 million to develop the stupidest thing he had ever seen.
Then again, the payoff for the fastest- sailor-in-the-world thing had been exactly $0.
It was enough to make Jenkins reevaluate.
Who exactly was the stupid one here?
“Poverty is the mother of invention,” Jenkins says, and then he shows me the tail tab control.
It’s just a piece of string connected to the tail by means of an antique fishing reel.
Jenkins, eager to let some hot air out of the Liquid Robotics balloon, originally planned to send Saildrone chasing after Wave Gliders as they crept across the Pacific.
When the Wave Gliders launched, Jenkins hadn’t even started building.
Still, he knew his boat was at least five times faster, and he calculated that he could beat the Wave Gliders to Australia even if he got a late start.
But by the time he was able to raise the necessary funds—largely through a grant from the Marine Science and Technology Foundation—the clock had run out.
After more than 300 days at sea, two Wave Gliders made it to Australia.
The other two broke down en route to Japan; only one was recovered.
Saildrone may not have had a chance to best Liquid Robotics on the water, but a head-to-head comparison of the two companies is no contest.
Liquid Robotics is seven years old, employs 110 people, and has now raised more than $77 million in venture capital.
The total bill for Saildrone, three years into the project, including what Jenkins and Owens pay themselves and their two helpers?
“Less than $400,000,” says Wendy Schmidt, wife of Google executive chair Eric Schmidt. (The couple funds the Marine Science and Technology Foundation.)

Schmidt regales me with stories of Jenkins’ incredible talent for dollar-stretching.
“He obtained massive amounts of carbon-fiber castoffs from his contacts in the yachting industry; he bought a broken $25,000 milling machine on craigslist for $2,000 and fixed it himself; he sublet an 800-square-foot workshop and then doubled it by building another floor!” Schmidt says.
“I mean, who does that?”

“Poverty is the mother of invention,” Jenkins says, and then he shows me the tail tab control—the throttle—inside the Greenbird’s cockpit.
It’s just a piece of string connected to the tail by means of an antique fishing reel.
“My grandfather’s,” he says.
“It was just cobbled together out of parts I had.”

 Richard Jenkins (right) designed the sailing robot; Dylan Owens handles the electronics.

Long way from Dead

The morning after the Honey Badger arrives in Hawaii, it’s time to send her out again.
The new mission is to spin the odometer past 7,939 nautical miles and thus rob Liquid Robotics of its endurance record.
Barefoot on the dock of the Kaneohe Yacht Club, Jenkins opens his iPad and drops a few new waypoints into the Honey Badger’s brain—aiming it around the South Pole and toward the equatorial Pacific.
If successful, it will be the first drone of any kind to “circumcise the world,” as Jenkins gleefully puts it.
It’s approximately 25,000 miles—10 times the distance to Hawaii—with no pit stops.
What are the odds?
“It’s a long shot,” says Jenkins, who points out that Saildrone was designed to get to Hawaii, no more.
“We cut a lot of corners when we started,” Owens agrees, “because we were paying for it out of our own pockets.” Hawaii was the milestone that Schmidt wanted.
“Every sailing journey that I’ve ever been on has been a near disaster,” says Jenkins, sipping his beer from the bottle. “We’ll see.”
A month later Jenkins and Owens are in their new shop in Alameda, California.
It’s cavernous—an acre of covered space on a decommissioned Navy base—and littered with all sorts of fun engineering projects in various stages of completion.
In a far corner is the Greenbird, now shod with skates and being prepped for an eventual run at the ice-sailing record.
Off to one side is the carbon-fiber shell of a seaplane that Jenkins is building.
Across one wall is the beginning of a Saildrone production line. Orders are starting to come in.
But dominating the workshop and in the center of everything is another wing, one far bigger than the 20-foot-high drone wing, bigger even than the Greenbirds’s 28-foot-high wing.
This one is the people mover.
The ferry wing is the culmination of six years of lobbying by Jay Gardner, a self-described “self-employed hippie” who operates a small boat-charter company out of San Francisco.
The winds that blow through the Golden Gate year-round, the 60-year-old Gardner says, “are the closest thing to a perpetual-motion machine we’ve got.”


Dominating the workshop is another wing, one far bigger than the Greenbirds’s 28-foot-high wing. This one is the people mover.
Armed with an engineering study that predicts that sails retrofitted onto Bay Area ferries would cut yearly fuel costs by 30 to 40 percent, Gardner managed to get seed funding for a real-world ferry-wing test.
When it came time to actually spend that money, the choices were to acquire and fix up a $10 million experimental ferry wing that the Navy had partially built or to hire Jenkins to build a scaled-up version of his Saildrone wing for pennies on the Navy’s dollar.
Going with Jenkins was not just a question of cost, however.
Thanks to the sail-with-a-tail design, his technology requires no knowledge of sailing to use.
“You can have a button that says ‘Turn off the wind assist,’” Gardner says.
The wing sail is 40 feet high, 10 feet across, and as thick as a man is wide, but its fuel-saving potential—anywhere from $500,000 to $1 million per ferry per year—must be verified by scientists from UC Berkeley’s Transportation Sustainability Research Center.
They will be rolling it out of the shop, standing it up, and using a crane to mount it on a passenger-carrying trimaran that will travel around the bay for three months.
“It’s going to be awesome,” Jenkins says, his face turning froggy.
“We’re two weeks away from an erection.”
The ferry wing is just getting its finishing touches when Jenkins announces to his staff of two—a boatbuilder and a sander—that it’s time to go to the pub for the traditional Friday afternoon pint.

Before Jenkins leaves the shed, he checks on the Honey Badger’s progress for the day.
The odometer stands at 6,000 nautical miles, but something looks wrong.
Digging into the data he realizes that the sensor that measures the rudder’s angle is sending random garbage to Saildrone’s brain. Salt water must have somehow infiltrated the connection.
“That was the last analog circuit on the boat prone to corrosion,” Jenkins says, cursing himself for not having upgraded it.
“The new version of the boat is all digital.”

Jenkins and Owens start sending commands to the drone and realize that all is not lost.
Even without the rudder, the wing should be able to steer it back to port.
“It’s a long way from dead,” Jenkins says.
Reviving the boat is a simple matter of swapping the old-style analog rudder encoder with the new-style digital encoder.
The only catch is that they’ll have to go to Hawaii to do it.
Another trip to Hawaii?
The crew greets the news with a chorus of clinking glasses: “Honey Badger don’t care.”
And then they raise their pints high for their traditional toast. “Honey Badger don’t give a shit!”

Links :

No comments:

Post a Comment