HD PBS / NOVA - Making Stuff Wilder（讓東西變得更狂野）。 (HD PBS / NOVA - Making Stuff Wilder)

字幕列表影片播放

DAVID POGUE: Civilization is built on the human drive to invent.
We take the raw stuff of our planet,
the materials that give names to the ages--
stone, bronze, iron, and more--
and craft them into new forms,
expanding our horizons...
exploring hidden worlds...
and engineering life-changing technologies,
always pushing the limits
to be colder, faster, safer, wilder.
And now a new era is upon us,
as scientists turn for inspiration
to the ultimate inventor and engineer: nature.
What can we learn from living things
to make our own technology even better?
How long before they become self-aware
and turn on their overlords?
POGUE: I'm David Pogue, and I am on a quest
for the world's wildest new stuff.
From a carnivorous tropical plant...
Whoa-- "Little Shop of Horrors."
POGUE: ...to an elephant's trunk...
A little elephant snot for you.
POGUE: ...there's a revolution underway,
as scientists borrow the best ideas nature has to offer...
I feel like an outtake from Ghostbusters.
POGUE: ...and put them to work,
creating a robot as big as an ox...
LS3, get up.
POGUE: ...and just as sturdy on its feet.
The power!
POGUE: Even teaching viruses...
I got... oh!
POGUE: ...to make batteries.
Come on!
POGUE: Nature has been making stuff
for billions of years.
What happens when scientists open up its toolbox
to make stuff wilder?
Major funding for NOVA is provided by the following...
We know why we're he
We know why we're here...
To chart a greener path in the air and in our factories.
MAN: To find cleaner,
more efficient ways to power flight.
And harness our technology for new energy solutions.
Around the globe, the people of Boeing are working together
to build a better tomorrow.
That's why we're here.
s.
And the Corporation for Public Broadcasting, and:
Major funding for "Making Stuff" is provided by:
Additional funding is provided by:
POGUE: We humans love to invent.
We've been doing it for thousands of years.
But how many of our inventions really stand the test of time?
Now imagine a world filled with only the very best stuff:
amazing ideas and astonishing designs,
each one the result not just of years, not of decades,
but millions of years of research and testing
in an environment where the competition can be ruthless.
Since life began on Earth, it has been innovating,
making discoveries in materials and engineering
we've only recently begun to appreciate.
The hard-won lessons of life on Earth, gained over eons,
may help solve our very human problems.
Can forms found in nature reshape our machines,
making them more useful?
Can we build the agile movements of animals into our robots?
Have some of the best ideas for new materials
already been discovered by nature?
What if we could make things like nature does?
Can we grow the electronics and fuels of tomorrow
using the code of life itself, DNA?
The search for answers to these questions
has taken some strange turns.
Here at the University of Guelph,
about an hour outside Toronto,
materials scientist Atsuko Negishi
and biologist Julia Herr
think that these lovely creatures called hagfish
may revolutionize how we make strong materials.
JULIA HERR: So these are Pacific hagfish.
They are well known for their unique defense mechanism.
POGUE: So if I wanted to see this, what would we do?
Like, could we poke it with a stick?
JULIA: I think the best way to do it
is to reach in there and grab one.
Oh, my gosh.
Look at that disgusting...
Oh, no!
I've been slimed!
I feel like an outtake from Ghostbusters.
Look at the quantities of this stuff!
This is like three times the volume of the fish.
How could all this slime come out of that tiny thing?
POGUE: It's an impressive display of slime... ocity,
and it works great as a defense.
When a gill-breathing predator bites down on a hagfish,
it gets a mouth full of slime.
With its gills clogged,
it becomes more worried about suffocating than eating.
So how does the hagfish conjure all that slime?
One of the key components is mucin,
a family of proteins that includes...
(loud sneeze)
...you guessed it: human mucus.
Mucin consists of a protein backbone,
with lots of sugar side chains hanging off of it,
like bristles on a brush.
These side chains attract water molecules,
soaking them up remarkably well--
in fact, amazingly well.
Atsuko offers to show me
what a little dab of hagfish juice can do.
This is a beaker of seawater,
so we're going to try to make some slime.
It's getting a little misty.
I'm just gonna mix it up a little bit.
All right.
And if you could lift that out.
Look at that!
Oh my gosh.
That tiny dab...
Hey, there's no water left!
It's taken the entire thing of water with it.
That little tiny pea's worth of white stuff...
Would anyone like some, children?
There's plenty to go around!
POGUE: While the mucin sucks up the water,
it's a second component that holds the slime together
so I can pick it up:
these threads, visible here.
Both components start out inside pores
along the side of the hagfish.
The water-loving mucin molecules
are packed into one type of specialized cell
while the threads are wound tightly in another kind of cell.
When under attack,
the hagfish ejects the cells and they break open.
The mucin molecules collect water molecules
while the threads, each about six inches long, unfurl,
binding the mucin into a continuous
and disgusting mass of slime.
It's these spider silk-like threads
that have really caught the researchers' attention.
They might even serve as a model for a new kind of fiber
because they're surprisingly strong--
ten times stronger than nylon,
a synthetic material made from petroleum.
If we could use hagfish fabric instead,
it could help reduce our dependence on oil.
So what are the steps involved in going from hagfish slime
to handsome garments made of it?
So the first step would be
to be able to artificially make these hagfish slime threads.
POGUE: It's early days, but Atsuko has been working on a process
to create her own fiber using proteins she's derived
from freeze-dried hagfish thread.
She mixes the proteins with formic acid
and puts a few drops onto a salt solution,
then draws up the material
to create her own artificial hagfish thread.
So far, it doesn't test as strong as the original,
but she has high hopes.
So you started with hagfish fiber.
You treated it to come up with this component goop,
put it back into saltwater,
turned it back into a piece of thread.
So you start with a thread, you ended with thread.
Why didn't you just use the thread to begin with?
One of the reasons is because we can't farm hagfish.
You can't farm hagfish?
They don't currently reproduce in captivity,
and so we can't have these big farms of hagfish.
I see.
Plus, it'd be a pain to get up at 4:00 in the morning
to go milk the eels.
POGUE: So all of this thread pulling is really in anticipation
of the day Atsuko can synthesize her own hagfish proteins.
There it is!
Actual thread made of actual reconstituted fish mucus.
The dawn of the era of hagfish fabric,
right there.
POGUE: Hm...
What would that be like?
Nighttime is the right time for a fabric from the deep:
Hagwear.
But it's look, don't touch...
(gasps)
...or the surprise will be on you.
Hagwear!
All right, hagfish fabric may not yet be runway-ready.
But in the right hands, nature's innovations offer clues
that can literally shape the stuff we make.
Built to thrive in their environments,
animal bodies offer winning designs and possible solutions
to our own engineering challenges.
After all, feathered wings inspired our metal ones.
Sleek swimmers helped shape our boats.
What other new solutions can be found
by studying the forms of animals?
I'm in Stuttgart, Germany, at the Wilhelma Zoo.
It might be the perfect place
to see the future shape of technology,
according to engineer Heinrich Frontzek.
So when an engineer like you comes to the zoo,
do you see it differently from regular visitors?
I think so,
because we want to get inspired by the nature,
and here in the zoo, they're so concentrated,
the huge variety of animals
all optimized for their applications,
and we are thinking in applications,
so this is a paradise for an engineer.
POGUE: Heinrich works for an automation company
trying to improve one of the most important inventions
of the 20th century: the robotic arm.
It's been revolutionizing factories since the first one
was introduced in 1961 at General Motors.
But robotic arms have some problems.
Just like the one on humans, the traditional robot arm
consists of rigid parts joined together,
often limiting its programmable motion.
They're also dangerous.
Get hit by one of these, and it's lights out.
So robots often end up behind protective fences,
unable to work closely with humans.
The German automation company Heinrich works for-- Festo--
decided to reinvent the robotic arm,
making it more flexible and less dangerous.
Heinrich leads me to the source of the inspiration,
and it turns out maybe the best arm is a nose.
So why would you look at an elephant's trunk
and think this would help you with automation?
As you can see, it's so flexible and transmits a lot of force
and makes it much more easier to handle things.
And this is our business, handling things.
To automate factory or a process,
and it make sense to look into nature
and to get inspired by nature,
and the elephant is an excellent ambassador for that.
POGUE: Thanks to Zella, a 47-year-old Asian elephant,
I get a little first-hand experience
with what an elephant packs in its trunk.
Little elephant snot for you.
POGUE: An elephant trunk is an impressive multi-tool,
able to slurp up water...
ZOOKEEPER: Now she collects the water.
POGUE: ...and squirt it.
Breakfast is on!
POGUE: It picks up food like a vacuum cleaner,
manipulates objects, and it's strong.
Zella can use her trunk to lift over 400 pounds.
No, no, that's my wrist.
She could crush me like a bug, couldn't she?
Yes, sir.
Here, have some more peanuts.
POGUE: But the trunk's most impressive attribute
is its amazing flexibility.
It comes from having no bones and about 40,000 muscles
arranged lengthwise and in rings.
With no bones and no joints,
it's about as far away from a traditional robotic arm
as you can get.
(beeping)
I head to Festo's headquarters in nearby Esslingen
to see their version of the elephant trunk.
They call it a bionic handling assistant.
Now, this looks like a bionic handling assistant.
Yeah, you're absolutely right.
This is our trunk.
POGUE: Festo's version of the trunk is made of plastic
with a series of air chambers inside.
Filling different parts with compressed air
causes it to bend.
So if I wanted to bend that way?
We need a tube with compressed air
for this expansion,
and then you get this bending to the other end.
So this blows up like a balloon?
Yes, for sure.
POGUE: They're testing the assistant with this simple motion
for use in a packaging operation.
Look at that, it tucks it in nicely.
Well done, Dumbo.
POGUE: But it is inherently more flexible
than a conventional arm,
and just as important, far safer.
We don't have electricity,
we don't have steel and iron and all this masses,
which could damage a person.
It's a weight of five pounds,
some valves, a little control system...
So there's really nothing here but plastic tubes and a.
Yeah.
Does it do tricks?
(laughing)
POGUE: In this application,
the tip of the trunk works by suction.
But Festo has experimented with what it calls a "fin gripper,"
inspired by fish fins.
If you push on the middle of a tail fin,
it doesn't curl away from you as you might expect;
it curls toward you,
giving a fish much more efficient strokes.
But Festo has built that principle into a gripper
that curls around the object it needs to pick up,
adapting to the shape.
So it looks to me like
you're about to demonstrate how this might work.
FRONTZEK: Yes, we have two different gripping devices:
one with a fish tail,
and this is a traditional one the robots are using.
Can I see these things close?
FRONTZEK: Sure.
Same pressure, everything is equal.
Now we will see what happens.
This is the old robot
and this is the bio-engineered method.
Okay.
Let the competition begin.
Look at that!
Traditional robot hand, shattered to smithereens,
and the fishtail gripper really did its job.
So you have robot zero, fish tail one.
You have stolen from nature and did a great job.
Thanks.
They'll edit all this out.
POGUE: Combining the fin gripper with the elephant trunk
produces a flexible, lightweight and safe robotic arm
ready for all sorts of applications.
FRONTZEK: Biomimicry nowadays,
it's part of the design process here at Festo:
cross-thinking, getting inspired by nature
and to transform these ideas into industrial applications.
May I?
Thank you.
POGUE: Festo's handling assistant
steals its form from the elephant trunk.
But Festo isn't alone
in adapting designs found in nature
and applying them to industry.
The beak of the kingfisher bird
breaks the water with very little resistance...
...inspiring the shape of this Japanese Bullet train
so it would cut efficiently through the air.
The shape of the yellow boxfish provides a rigid structure
and has very little drag for such a large volume,
all reasons Mercedes Benz used it
for the design of a high- efficiency concept car.
But making machines that look like animals is one thing.
What about making machines that move like them?
For thousands of years,
when we've invented new forms of transportation,
many have been based on a human insight
not found in nature at all: the wheel.
But there are plenty of places wheels can't go,
even ones wearing a belt of tank tread.
The inability of our machines to traverse difficult terrain
has dire consequences in the battlefield
and in search and rescue.
But while wheeled vehicles struggle off road,
there are some creatures getting around on legs.
That's had engineers wondering:
what lessons can we learn from animal movement?
Can we give our machines a leg up?
Walking is easy for animals.
Even a toddler can do it,
Excuse me, Sir!
POGUE: And thanks to movies,
creating walking machines seems easy too.
Just look at C3PO from Star Wars...
Oh, nice to see a familiar face!
E chu ta.
How rude!
POGUE: ...or its Walkers.
You'd think the problem's been solved,
but in real life, it's hard.
One of the best-known early attempts at a walking machine
is General Electric's walking truck from the '60s.
It even tackled uneven terrain,
but it took a human operator to decide
where to place each foot one at a time,
an exhausting task.
By the '70s, computer control automated the walking motion
in a series of crawlers built around the world,
though still driven by human operators.
These kept a tripod of legs on the ground,
maintaining stability at all times:
a system called static balance.
They moved slowly, like a walking table.
But in the 1980s,
a very different approach gained ground.
I've traveled to Massachusetts
to visit a company that builds robots based on that work.
The company's founder, Marc Raibert,
has been building walking robots for over 30 years.
Early on, he steered away from the static balance
of walking tables.
To help me understand how he views animal locomotion,
he invites me to take a ride...
...on a pogo stick.
It's tricky because like all standing humans,
I am top heavy--
in technical terms, an inverted pendulum.
Here is a normal pendulum, right?
If you swing it, it just hangs down.
MBut if you put the weight at the top, what happens?
If you don't do anything, it tips over.
But if you move the point of support,
you can keep it balanced.
POGUE: If you're top heavy, staying balanced requires
keeping your base of support under your center of gravity.
That's what Marc is doing
by shifting the bottom of the broom as it tips,
that's what I'm doing by moving around the pogo stick,
and that's what all of us do all the time when we're upright.
In fact, the human brain receives constant updates
to maintain the body's balance:
from the inner ear, where a series of fluid-filled canals
send signals about the position and motion of the head;
from the eyes, which send signals
about the body's position relative to other objects;
from internal sensors that tell us
about the position of body parts relative to each other;
and from external pressure sensors in the hands and feet
that send signals about the source of support--
for example, if you're on uneven ground.
All of this information feeds into our cerebellums,
which keep our top-heavy bodies from tipping over,
even when we're just standing around.
Without it, you would topple over.
To Marc, we are less like a static table
and more like a pogo stick.
To focus on the problem,
he built a robot that had only one springy leg.
It constantly calculated
where its weight needed to shift to stay upright.
Very pogo-stick like.
Even when he added more legs,
he kept the bounce in their step
and an active sense of balance.
For the last few years,
Marc has been applying what he's learned
to solve a problem for the U.S. military.
On rough terrain, wheeled vehicles aren't much use.
And soldiers often haul everything on their backs,
leading to injuries and exhaustion.
Marc invites me to see Boston Dynamics' solution
out at a nearby park.
Meet LS3, also known as AlphaDog.
It's designed specifically for rough terrain--
anywhere a soldier might go on foot--
and it carries 400 pounds of gear
along with enough fuel for a 20-mile mission.
So in this mode, the robot is following the leader.
He's got a backpack on
that has some reflective stripes on it,
and the vision system focuses on that,
and then it records what path he takes through the terrain.
POGUE: Is it modeled after a particular animal-- an ox or a horse?
RAIBERT: Not really.
We take inspiration from how animals are designed,
but then we have to use human engineering tools
and human materials,
so sometimes it stays like the animals,
sometimes it departs.
POGUE: I ask Marc for a tour of LS3--
of course, after it's been shut off.
RAIBERT: So this is the leg and it's got a muscle her,
or the actuator, which causes it to move.
This muscle moves the knee joint.
The computer is really the equivalent
of a laptop-style computer.
So you know, all the balancing is done in that computer,
all the speed control, all the turning.
There is a laser range finder here
that provides 3D-depth information.
There is a set of cameras here
that look right in front of the robot
and provide information about the shape of the terrain
so that the feet can pick the best places to step.
The legs themselves can feel the forces
that are exerted at all the actuators.
POGUE: Does it ever slip?
What if it steps on an oily leaf or something?
RAIBERT: It slips, and frequently, it corrects for those slips.
So the goal is to make it so that the feet can slip
and the control system recognizes that it's slipping
and compensates by using the other legs.
POGUE: Lots of cool tech on LS3, but my favorite feature?
Voice control.
"Power on," "engine off," "sit," "get up,"
and "get me a beer."
That's a good one, I like that.
LS3, get up.
The power!
LS3, follow tight.
What a good boy!
LS3, sit.
LS3, power off.
(engine shuts off)
I think you got something here.
Nice!
POGUE: In a final test of LS3,
Marc has it "follow the leader" up a steep incline.
Here, you really see it actively balancing while in motion,
just like me on the pogo stick
instead of moving from one stable position to another
like a walking table.
RAIBERT: And the idea that you could have it
passively stabilized like a table, you know,
that doesn't really work with a moving robot.
There is too much energy in the motion of the body.
I believe that the only way to make these things work
is to really commit to the active balance.
POGUE: LS3 isn't really built for speed.
It trots at about five miles an hour.
But what would it take to make it go faster?
This is the Cheetah.
Just like the real deal,
its back flexes with each step,
increasing the stride of its gallop.
Right now, it's the fastest robot with legs in the world.
But start looking over your shoulder
for the next generation:
Wild Cat.
It's designed to be untethered.
(sirens)
POGUE: Four-legged robots have their uses,
but events like the recent Fukushima nuclear disaster
have renewed interest in the human form.
Radiation kept people at bay,
away from all available rescue equipment--
from cars, to power tools, to shut-off valves.
But imagine if there'd been
an easily controlled humanoid robot to operate them.
Robotics engineers have been working on that for years.
In 2009, Boston Dynamics introduced PETMAN,
a robot that balanced itself, walked,
and even did some calisthenics.
Over the last few years, PETMAN has evolved into...
Atlas, which has even more mobility.
Just like LS3, it actively balances itself all the time.
And in this impressive demo, all by itself, it uses its arms
to work its way past a hole in the floor.
Today they're tweaking its sense of balance on one foot.
Looking at what test to do here, we studied gymnasts.
And when they are just about to fall off,
you'll notice that they throw their arms and their legs around
very violently.
So we're trying to understand what techniques they're using
to build a robot that can really handle rough terrain.
POGUE: They've been doing this test for only a week.
First the robot goes up onto one foot.
Then they hit it with a 20-pound medicine ball.
PLAYTER: So, if you notice there, it's swinging its arms and legs
all around in kind of a clockwise fashion
and that momentum helps move the center of mass
back over the feet.
Not dissimilar to a way the gymnasts do it.
Whoa...
Now let's see some human dynamic balancing.
PLAYTER: The robot's blind.
It doesn't know the ball is coming.
Oh...
So, we don't want you to know the ball is coming either.
So, we've got a little blinder there for you,
so you don't see the ball coming.
Oh, great.
So I don't know when the ball is coming.
I have a feeling
if your stinking hunk of silicon and hydraulics
can do it, I can of course do it, too.
That's right.
POGUE: Side by side, it's hard to say who does it better.
The Atlas seems more stable.
But I have a few other tricks up my sleeve.
Very good.
I admire your robot, sir.
Well, I admire you wearing those glasses on public television.
(laughing)
POGUE: We'll be seeing more of this guy.
Atlas is the hardware
used by seven software development teams
in an upcoming international rescue-robot competition.
But a single sophisticated and expensive robot like Atlas
is just one strategy.
What about a less expensive and less complex machine,
but more of them?
That's the idea behind Harvard University's Robobee.
It would take 30 of these to equal the weight of a penny.
What happens when you move beyond having just one robot
and instead have a swarm?
In the future, swarms of robots operating as a team
might build our skyscrapers or map uncharted areas
or scout out victims in disasters
as robotic search-and-rescue teams.
But in order to do any of that, engineers must solve a problem
nature solved eons ago:
How do you get a group of individuals
to work together as one?
In nature, swarms often behave
as if they have a collective intelligence.
Whether it is fish schooling in the sea
or birds flying in a flock, the members act in unison
without anyone apparently in charge.
Some of the achievements built out of this swarm intelligence
are awe-inspiring,
like this murmuration by thousands of starlings...
or these complicated towers built several feet high
by blind termites.
So what can we learn from behavior in nature
about creating robotic swarms?
Vijay Kumar and his students at University of Pennsylvania
have been wrestling with the problem.
They use a fleet of hand-sized quad-rotor robots
which they've learned to manipulate
with impressive control.
They can play the theme from James Bond...
or put on a light show.
In both performances,
the quad rotors are individually controlled
by a central computer.
But they've also built some computing power
into individual robots, so they can think for themselves..
like figuring out how and when to fly through a tossed hoo.
Now Vijay is taking the next step, developing software
that will allow the 'bots to work together as a swarm,
a team that can do more than any single flyer can.
One flying 'bot, pretty cool.
Eight flying 'bots?
It gets a little swarm in here.
KUMAR: So what you see here
is these robots are commanded to rise into a swarm.
They're asked to form patterns, three-dimensional pattern,
and then the robots figure out what point in the pattern
to step into and how to coordinate with their neighbors.
POGUE: Oh, so the master computer doesn't say,
"You be in the corner."
It's just saying, "Be a rectangle," "Be a circle,"
but they have to decide how to execute that?
KUMAR: Right.
POGUE: While a central computer
could control each of the eight robots individually,
telling them where to go,
Vijay wants a system that scales up.
And with more robots, no computer could keep up.
So instead,
he's taken inspiration from swarms in nature
and developed three guiding principles.
First, as much as possible, just as in nature,
each robot thinks for itself.
Second, each robot acts primarily
on local information it gathers,
the way a bird in a flying flock probably pays attention
only to its immediate neighbors to know where to go.
Finally, no one robot is in charge.
They're all interchangeable.
So that if one breaks down, the group continues.
To test out those principles,
Vijay turns his fleet over to me and lets me experiment.
The flyers know they're supposed to make a circle.
As I add them one at time, you can see it take shape.