La Ciencia del Futuro - Stephen Hawking. Hyperconexiones
En este episodio, Stephen Hawking y su equipo investigan el amanecer de una nueva era en la que las máquinas hablarán con las máquinas, en la que ciencias previamente aisladas se conectarán para formar nuevas disciplinas, y en la que, incluso a nivel atómico, se forman cada día nuevas conexiones.
Category
😹
FunTranscript
00:00I believe that one of the characteristics that will define this decade will be connected technology.
00:10Machines are connecting and different disciplines and industries are coming together.
00:15Hyperconnections are allowing us to make revolutionary discoveries and even discover new scientific fields.
00:22By sharing ideas, new and incredible advances suddenly become feasible.
00:31Hyperconnections
00:34Even the greatest machines we have created in history need us to assign tasks to them and make them work.
00:43The problem is that most of them can not make reasonable decisions like us.
00:48That is, they are basically stupid.
00:51But connecting machines together allows them to gather information about the world around them and share it.
00:58An open door to new possibilities in research to create a true artificial intelligence.
01:09Jim Al-Khalili is in Zurich, Switzerland, to investigate the innovative and revolutionary technology
01:15that allows machines to teach each other and make their own decisions.
01:21The scientists of the Zurich Federal Polytechnic School are testing an innovative concept
01:26known as a smart feedback loop in a new generation of machines.
01:31Something never seen before.
01:35These quadcopters work thanks to one of the most extraordinary artificial intelligence connections in the world.
01:42Quadcopter
01:47At this time, four mini helicopters are preparing for a test of obstacles.
01:52We have already put the slalom bars. What are the quadcopters going to do?
01:56They will fly a couple of times over the bars to learn the route, just like an Olympic slalom skier would.
02:03You have to know it first.
02:05He will do it by repetition. He will go through the circuit a few times,
02:09and once he is able to perform the slalom properly, he will descend and do it at full speed.
02:20Equipped with the latest technology in microprocessors,
02:23each quadcopter is programmed with an exceptional learning algorithm connected to its central computer.
02:30Here on the screen you can see the path it is traveling.
02:33Exactly. Here we can see that if it were flying through the bars, it would be colliding with them.
02:38As it flies through the different obstacles, it records its position 50 times per second
02:44and sends that information to the algorithms of these computers,
02:47and at the end of each attempt, it checks which positions it has been in and where it should have been,
02:52and in the next attempt, it compensates for the error.
02:55This is the smart feedback loop in action.
02:58Once you have learned the route, it is time to put the technology to the test.
03:04Now it's dropped. It looks like it's ready.
03:06Yes, it's ready, Jim.
03:17Perfect. Here you can see the route. It has learned it.
03:20It has gone through all the bars without touching them.
03:22Once the first machine has passed the challenge,
03:24it can be transferred to other machines that have not faced it.
03:28The new machines, without testing the circuit,
03:30would do it correctly the first time without making mistakes.
03:35It's impressive.
03:36These quadcopters are not only able to learn from their experience,
03:40but they can share it instantly with all the others.
03:47The process is very fast.
03:49They are able to learn to perform tasks of great complexity much faster than we humans would do.
03:55By updating their decisions based on the decisions of other machines.
03:59But how will this technology be given such a complex task as balancing an object?
04:04Dr. Rafael Odandrea is the leader of the project.
04:08Jim, I have a mission for you. Try to keep this stick upright with your finger.
04:12Okay. Normally, I'm good at these things.
04:16It is a challenge for anyone who is moderately coordinated.
04:20Of course, it requires a lot of coordination between sight and hand.
04:23Right now, three things are happening.
04:25First, you have to look at the stick,
04:27you have to process the information about what you have to do,
04:30and then you have to move your hand so that the stick does not fall.
04:34Yes, there are quite a few processes.
04:42The truth is that the quadcopters kept the stick in balance better than me,
04:46but would they be able to do something that for me would be totally impossible?
04:50Could this smart feedback loop make them able to throw the stick at each other?
04:59Wow, this is amazing. But how do they do it?
05:02On the roof, we have eight cameras mounted,
05:04which act as a kind of closed-door GPS system.
05:07The information is sent to various computers,
05:09and the orders are then sent to the quadcopters at a frequency of 50 times per second.
05:14The quadcopters then execute those orders internally a thousand times per second,
05:18so that the entire process takes about 20 milliseconds to develop,
05:22from the perception and sending of the information until the machines decide what to do.
05:26Ten times faster than people then, right?
05:28Exactly, at least ten times faster, and sometimes up to a hundred.
05:31But the most remarkable thing is that it is not remotely controlled by a human operator.
05:36That's right, it's a completely autonomous system.
05:39In fact, if a person would find it impossible to fly the quadcopter while keeping the stick in balance,
05:44imagine if he had to throw it up in the air so that another quadcopter would be below him and catch it.
05:53Wow!
05:55These machines were able to pass a stick from one to another,
05:58but what if we introduced an unforeseen factor?
06:01Something they couldn't control.
06:03In the next test, these machines, able to process and send information at high speed,
06:08will have to calculate the trajectory of an object that will be thrown at them randomly
06:12and make decisions about how to act.
06:14Well, here you have Jim.
06:15Go ahead, throw the ball hard into the area where the quadcopters are.
06:18They will have to be able to intercept it, to catch it.
06:20Okay.
06:21Very good.
06:22Wow!
06:23Very good.
06:24What they will do now is get closer and try to get it back to you.
06:26Yes?
06:27Get it back to me, right?
06:28Very good.
06:30All right, good.
06:31What is demonstrated with this?
06:33So what you've seen is obviously not something that a single quadcopter would be able to do.
06:37So they're working together to do a dynamic task together.
06:41They have to work together, otherwise the network could get caught.
06:45Each one has to pull it in a specific direction.
06:50And the quadcopters go one step further.
06:53They are able to throw the ball from one to another with incredible precision.
07:01As a concept, it is impressive.
07:04Machines capable of making decisions faster and with more precision than us.
07:10Is this finally the beginning of a new era?
07:15When do you think this kind of technology can start to be part of our lives?
07:21I mean, so long they've been acquiring relevance without a break.
07:24But you'll see how the process accelerates now that we can share all this information.
07:28I think we'll see how the effect of this mechanism increases the possibilities of these machines.
07:34So we're talking within a few years.
07:36I think so.
07:40In 2018, Raffaello believes that this technology has already been fully developed and that it will be working all over the world.
07:48Even the most conservative estimates predict that at least 7,500 devices similar to these will fly over the United States.
07:56The promise of having connected machines among them is one of the tangible examples of a hyper-connected world.
08:01A world in which machines learn from each other and collaborate without human intervention.
08:07You know, it's an impressive advance in artificial intelligence.
08:12These machines are much more intelligent than I expected.
08:15You know, the truth is that what I've seen here today has really impressed me.
08:18They're not simple intelligent toys.
08:20However, if there's one thing that's guaranteed, it's that this technology can only become more intelligent.
08:28Now, as more machines develop their ability to work together,
08:34people will stop directing them to supervise them.
08:39I can imagine countless ways to drastically change our day-to-day with a technology of this type.
08:45Helicopters that distribute mail.
08:47Unmanned truck fleets.
08:49Waiters.
08:50Surgeon assistants.
08:52All of this will be possible, and I think inevitable, by the end of this decade.
08:58Innovations are being developed using connections that operate on a microscopic scale.
09:05How can a small and popular creature that lives in the oceans
09:09go from serving us dinner to saving millions of lives every year?
09:17One of the most abundant creatures in the ocean has some hidden properties
09:22that could save millions of lives every year.
09:26The shrimp exoskeleton contains a natural coagulant agent.
09:31Daniel Kratz has gone to find out how scientists are using it
09:35to create a new blood coagulant agent
09:38by making several connections on a molecular scale.
09:49Every year in the United States, 30,000 people lose their lives in traffic accidents.
09:54Many of these deaths are caused by the time emergency services arrive
09:58and take the victims to the hospital.
10:02I am at the University of Maryland, on the outskirts of the city of Washington,
10:06to meet with a team that is developing a technology from nature
10:10and optimizing it to offer new solutions to hemorrhages and traumas.
10:15The crustacean exoskeletons began to be used in human healing in the 1980s,
10:19but this laboratory is going one step further.
10:21The developers are two biological engineers, Matt Dowling and Professor Srinvasa Raghavan.
10:26Hi, Srinvasa Raghavan.
10:28Nice to meet you.
10:29We started talking to some doctors from the Faculty of Medicine in Baltimore
10:32and they told us that the current hemostatic deposits designed to stop hemorrhages
10:36were not effective.
10:38And the inspiration came to us a few years ago, when Matt joined the team.
10:41Matt and Srinvasa study the gelatinous properties in nanoscale of polymers.
10:46Then we said, let's go further.
10:48We have a material that works in cellular structures.
10:51Could it also work in blood?
10:53The exoskeletons produce a material called chitosan.
10:56These biological engineers have found a way to improve their natural properties
11:01by adding small filaments that create a nanoscopic mass
11:04that, when mixed with blood, forms a coagulant.
11:09Is this your laboratory?
11:11Yes, this is our laboratory.
11:13And this is chitosan, the main basis of Hemogrip.
11:17What we are going to do now is to take the same chitosan powder that I have taught you
11:21and we are going to dissolve it in a glass to see what happens.
11:24After that, we will add a small amount of acid.
11:27Chitosan needs a little acid to dissolve in water.
11:31As it takes a while to completely dissolve,
11:33we have already prepared another glass with dissolved chitosan.
11:36To this we will add a few reagents that will transform normal chitosan into Hemogrip.
11:42Now we are adding fat molecules to the structure of the chitosan.
11:47This allows it to self-assemble,
11:49to form a physical seal when a biological fluid comes into contact with the blood.
11:54The fat waits until it comes into contact with the blood.
11:57When that happens, the altered chitosan comes into action
12:00and forms a biological precinct as resistant as a plastic film.
12:05Great, so this is your magic combination.
12:07Basically, you have a bottled biological deposit.
12:10Can we see it in action?
12:11Sure, we'll see it right now.
12:14Well, Daniel, let's take a look at normal chitosan first
12:17and its coagulant properties.
12:20Ian, can you add a little blood to the chitosan?
12:23Let's put it there, in a solution with normal chitosan.
12:27We mix it for a second and then we will reverse the route.
12:30And now we see what happens when it mixes with the blood.
12:33It remains like a viscous liquid that flows without problems.
12:36However, when we add the same amount of blood
12:38to the same volume and the same amount of chitosan hemogrip,
12:41we see that it quickly forms a gel that supports its own weight
12:44even when the route is reversed.
12:48Look at it yourself.
12:50In this case, the altered chitosan coagulated the blood instantly.
12:53The chitosan hemogrip is capable of making the blood
12:56become a million times more viscous than in its natural state.
13:01Arterial hemorrhages are currently the main cause of death for soldiers.
13:06Many victims would survive if there was a faster and more effective way to stop the hemorrhage.
13:12New technologies like this are urgently needed to save lives.
13:17Daniel, now I'm going to show you the uses that will be given to hemogrip in the field.
13:20The first of them is the deposit that we create in the test bench,
13:24which, after freezing and drying, becomes a material like this,
13:28which can be used for war wounds or for traffic accidents
13:31and even in surgical interventions.
13:33It is incredible to me that this can be put directly on a wound.
13:36It would give rise to a kind of mixture that would immediately stop the hemorrhage.
13:39Exactly. It has very strong mechanical properties that allow it to form deposits like these.
13:44Thirteen million civilians suffer from severe traumas every year around the world.
13:48Matt has developed a version of the chitosan that could be very useful in everyday emergencies.
13:56We have developed a foam format, a hemogrip spray in foam.
14:00Why don't you try applying it on the table?
14:04It is a foam capable of spreading in spray,
14:07which can be used in irregularly shaped wounds and in surgical wounds of people with coagulation disorders.
14:13But creating the deposit is only half the process.
14:16Matt has also devised a clever way to remove it.
14:20Well, let's say a patient arrives at the hospital with hemogrip covering the entire wound.
14:25The traumatologist will need to be able to remove it quickly and easily
14:29to be able to identify the injuries and operate without impediments.
14:33So, what we would have to do is add this reagent to the gel and then just give it a little mix.
14:43Incredible. Wow.
14:44So, using this agent, we invert its effect and we have liquid blood again.
14:48We have the hemorrhage under control.
14:50That's right. We can remove the hemogrip easily and effectively.
14:53Now that you can invert the hemogrip, it is incredibly useful for the surgeon,
14:57for the emergency doctor, who can see where the hemorrhage comes from,
15:00to operate and fix the problem.
15:04During the trip to the hospital, the patients are especially vulnerable.
15:08I wonder what a reputed traumatologist will think about this new discovery.
15:13In what contexts do you think this could be applied?
15:15Could it finally satisfy the needs of the traumatological sector?
15:18Some people think that our system of applying pressure and using machines like this is archaic
15:23and this new invention and the ease with which it is applied is a revolution for us
15:27because it completely cuts the hemorrhage.
15:30It should be noted that hemorrhage is the main cause of death in people with traumatic injuries.
15:34We hope that this will have an effect in many areas, that we can have it even in our kitchen.
15:39So, if someone cuts themselves with a knife, with a stabbing object,
15:42and a hemorrhage occurs, instead of applying pressure,
15:45this can be applied and stop bleeding until you get to the hospital.
15:49Then there is a great need to cover, Matt.
15:52When could it go on the market?
15:54Well, we hope that Hemogrip technology can be available in a matter of a few years.
15:59The chances of surviving an accident are increasing
16:03thanks to scientists who create molecular connections
16:06that make up the part of the crustaceans that we have left
16:09when we eat them in valuable medical products.
16:14Scientists from very different fields work together
16:17connecting their research in such a way
16:19that they could create a new kind of flying machine.
16:22One that is independent, smart, and above all,
16:26that flies as agile as a bird.
16:30Chris Elias Smith is going to find out
16:33if it is possible for pigeons to hide the secret of flying without a crew.
16:43There are thousands of pigeons in every city in the world.
16:47They are carriers of diseases, they damage our properties,
16:51they are considered winged rodents.
16:54Poor pigeons have a pretty bad reputation.
16:57I'm going to Harvard, in Cambridge, Massachusetts,
17:00where biologists and engineers collaborate
17:03on the connection between pigeons and planes of the future.
17:09First stop, the biologist Ivo Ross.
17:13I'm interested in how animals move,
17:16and the flight of pigeons is a form of movement,
17:19a movement that I find fascinating and that I wanted to know more about.
17:23Pigeons have been studied for years.
17:26Many investigations have been carried out
17:28on their motor skills and neurology, for example,
17:31and Ivo has used them to discover how they can fly with such agility.
17:36What we want to do now is combine all this
17:39to see how they use the tracks that give them some sense,
17:42like the sight, to control their flight.
17:44Do you have any of those little bugs here?
17:46Here we have a pigeon with a good tail,
17:48and this is the swinging movement.
17:50If I tilt it quickly in this direction, you see how the tail goes up.
17:53Yes, look what it does.
17:54It extends and raises it.
17:55Something similar happens with the wings.
17:57If I wanted to rotate one of them, the bird's reaction would be to extend it.
18:00Ivo's discovery occurred when he observed how perfectly still
18:03he kept his head and began to theorize
18:05that it was possible that this was related
18:07to how they see the world.
18:10So this is where you release it?
18:12Yes, we have a perch here on one side,
18:14and we release it so that it flies through these obstacles to the other side,
18:18and we study the decisions it makes.
18:20Now it's going to fly.
18:21There we go.
18:22That's it.
18:25It's very interesting that when the bird lands,
18:27it prepares the body, changes the position of the wings,
18:30and always lands on the perch without making a mistake.
18:32Yes, its senses provide a lot of information,
18:34and that's what we want to investigate.
18:36They also make a particular wing movement
18:38that consists of raising it,
18:40and we think that with this movement they create
18:42an aerodynamic force even when landing.
18:44To reduce the speed.
18:45To reduce the speed in particular.
18:46Yes.
18:47And to stay in the air.
18:48So can we see this bird do that too?
18:50We can put a few bars and see if the bird is able to sort them.
18:53Great.
18:54So one thing that we're very interested in
18:56is knowing if they're looking ahead and planning their journey
18:59based on their vision instead of resorting to previous knowledge
19:02about the configuration of the forest.
19:04If the routes that we propose here were very simple,
19:06it seems that they make their decisions at the last moment,
19:09when they are a meter and a half from the forest
19:11and start moving through the widest gaps
19:13that they see at that distance.
19:16This particular movement time is incredibly complicated.
19:19Yes.
19:20How do you analyze it and understand what's happening?
19:22That would be the next step.
19:24We want to know how they move, but in addition to that,
19:26we also want to know how they implement the control
19:28to execute these movements.
19:30Yeah.
19:31And that's what leads us to carry out
19:33a series of experiments in which we pay attention
19:35to what kind of signals we could use
19:37to control their flight trajectories
19:39and the orientation of their body.
19:41Ivo took me to the lab where the rigid theory is put to the test.
19:48What we do here is put small diodes to the animal,
19:51some small lights, these little batteries here,
19:54and we also use a small harness that we put on the bird and...
19:57A bird backpack.
19:58Yes, a bird backpack.
20:00And why do we put this backpack on it?
20:02The backpack has some small LEDs,
20:04some small lights that will shine
20:06and will allow us to use our program
20:08to track those markers
20:10so that we do not need to click on each of the frames.
20:13I understand.
20:14So now we have to try to find its spine
20:17to put the markers symmetrically
20:19and make things easier.
20:24In my life, I had imagined putting a backpack on a pigeon.
20:28So we have three on the body and one on each wing,
20:31and now we're going to put a little stick on its head
20:33and a LED on each side,
20:34which will show us the orientation of its head.
20:36I see.
20:37Knowing where the pigeon is looking is what interests us.
20:41Yeah, just set it up and it'll probably fly.
20:43That's it.
20:44It's gone to the other side.
20:45There it goes.
20:46Perfect.
20:47Could you explain to me what this is exactly?
20:49Yeah, basically it's a huge vertical cylinder
20:51with two perches at the top.
20:53What we're trying to do is make it fly in the dark,
20:55so we put the pigeon in here,
20:57we make it fly vertically,
20:59and then we rotate these vertical bars.
21:01Okay.
21:02At that point, we make it believe that the world is spinning,
21:05and what we want to find out is if the pigeon would spin with the world.
21:08So the backpack we put on the pigeon
21:10helps you make the measurements?
21:12Yeah, we just need the LED lights we put on the bird
21:14to capture it with the two cameras we're recording from above.
21:17Yeah.
21:18And then we reconstruct the trajectory in 3D.
21:20Great.
21:21Then we activate it so it rotates.
21:23We release the pigeon, and once it's up,
21:25we press the button, we pick up the information, and that's it.
21:28So this also makes the cameras stop recording
21:30and store the information we've got?
21:32Exactly.
21:33Okay.
21:34That's it. Okay, let's go.
21:35Okay, I'm ready.
21:36The world is spinning.
21:37We put the pigeon in,
21:38we press the button,
21:39and trigger.
21:40Perfect.
21:42Okay, so now we've got the information,
21:44and we've got the pigeon,
21:45so we're going to turn on the light,
21:46and see what we've got.
21:50Yeah, it's better with the bird in the hand.
21:52We press play, and see what we've got.
21:55Is this what we just recorded?
21:57What we see here is the individual points
21:59that follow the movement of the bird
22:01as it ascends through the tunnel.
22:03It's great.
22:04It's fascinating to see it at this speed,
22:06because it actually does it very quickly.
22:08They stabilize the head very well when flying,
22:10and that's something we're very interested in.
22:12It had already been proven previously
22:14that the pigeons stabilized the head
22:16and moved it like this before landing,
22:18but now we've discovered that they also do it
22:20in mid-flight, when turning, and even in slow flight.
22:28A team of engineers from the Massachusetts Institute of Technology
22:31is very keen to follow up on Ivo's research.
22:34Is it possible that the pigeons serve as an inspiration
22:37for an advanced robotic airplane
22:39to be able to fly like a real bird?
22:43Now it's getting really interesting.
22:45I've previously built brains,
22:47but linking them to artificial bodies
22:49presents a whole range of new challenges.
22:51Reproducing the fluidity of natural movement
22:53in a mechanical apparatus is not an easy task.
22:56A few years ago, we asked ourselves,
22:58is it possible to make a kind of robotic bird?
23:00This was one of the first mechanical birds available.
23:03There are a few more now.
23:05We affectionately call this one Phoenix.
23:07We know that birds are capable of doing spectacular things,
23:10like crossing forests at high speed.
23:12Until now, we didn't have much information
23:14about how they did it, and the data Ivo provides
23:16are the first to allow us, the engineers,
23:18to discover how they do such amazing things.
23:20Could I see how it works? I'd love to see it.
23:22Sure. Let's go.
23:24Andy has a joystick that controls the tail motors.
23:27This one here only controls the tail.
23:29It moves it in both directions.
23:31A new attempt to reproduce the control
23:33that we find in birds, right?
23:35Exactly. Would you like to take it?
23:37Okay, hold it.
23:38Yeah.
23:39There you go. Let's grab it.
23:41Careful, don't put your fingers in the gears.
23:43Gears?
23:44Yes.
23:45Wow.
23:46My goodness, it looks like it's going to take me flying.
23:49Unfortunately, it won't work that way.
23:51That would be awesome.
23:52So it can fly, really?
23:54That was its first autonomous flight.
23:56Yes.
23:57In the outside?
23:58Yes, in the campus of the Institute of Technology.
24:00Its system controlled everything.
24:02Wow.
24:03Until we decided it would crash into the building,
24:05so we decided to have it crash into the containers.
24:07Here we have another plane
24:09belonging to the next line of aircraft that we create.
24:11At that moment, we asked ourselves,
24:13well, what can a bird do that a plane can't do
24:15and that we could study?
24:16And we thought that finding out
24:18how they can land on a perch would be a good option.
24:21So, Joe, you made this plane?
24:23Yes, exactly.
24:24What are its main components?
24:26Basically, what we have here is a magnetometer sensor
24:29that perceives our magnetic field.
24:31We have a kind of accelerometers and gyroscopes
24:34that would give us the position of the plane,
24:36and we also use advanced techniques
24:38that, based on the measurements of the magnetic field,
24:40allow us to discover what position the plane is in.
24:42So those are the brains that control its body?
24:44Yes.
24:45Amazing. I'd love to see it in action.
24:47They took me to the motion capture laboratory.
24:50Right.
24:51So how does it decide to do a particular maneuver?
24:54Basically, this is a base station
24:56that is able to read the measurements
24:58taken by our motion capture cameras,
25:01identify the position where the plane is,
25:03and then execute a control algorithm
25:06that determines what kind of movements
25:08the depth rudder must make to land on the perch.
25:11Yep. Cool.
25:13The plane lands just like a bird.
25:15Getting it has cost the engineers a lot of work.
25:18As it comes off the launcher,
25:20it takes a picture of how fast it's going.
25:23They calculate the speed at which it goes
25:25and exactly what it has to do
25:27in order to lift the nose and land exactly on the perch.
25:30As you can see, the control system is deciding...
25:32Wow, that's amazing.
25:34...what the depth rudder has to do.
25:36Right.
25:37Basically, it uses the knowledge
25:39that the engineers have obtained
25:41from investigating the pigeons,
25:43based on what is seen at the last moment
25:45to make the correct avionics decision.
25:47Ivo set up some bars,
25:49and the pigeons had to go through that forest of bars.
25:52And we said, wow, that's very complicated.
25:54So what we were thinking of doing
25:56was a little bit more simple,
25:58putting two bars and trying to get the plane
26:00to go through the middle of them,
26:02but putting them closer together
26:04so that the gap is narrower than the width of the plane.
26:07So even if it's perfectly aligned,
26:09it can't go through in the middle like that,
26:11it has to do something, turn quickly.
26:13Right.
26:14In six consecutive flights,
26:16we've put two bars on each of them.
26:18What kind of use do you think
26:20could be given to this kind of technology
26:22once it's more advanced?
26:24I imagine a lot of short-term applications.
26:26We're building unmanned aerial vehicles
26:28capable of getting to places
26:30that you can't get to now.
26:32They could go through cities,
26:34get to buildings on fire,
26:36fight forest fires,
26:38places where we'd like to have unmanned planes
26:40and where they're not yet.
26:42Robotic pigeons, soon in your city.
26:44They will fly through the skies in 2020.
26:46But what else could biology
26:48contribute to engineering?
26:50As we unravel the mystery
26:52of bird navigation
26:54and the aerodynamics of their wings,
26:56will we see a completely different type of plane?
26:58Could we launch it again?
27:00Sure.
27:02The next generation
27:04of unmanned aerial vehicles
27:06will be able to fly hundreds of kilometers
27:08without the need for humans to guide them.
27:10If the connections between the birds
27:12and the planes
27:14are helping to define
27:16the future of unmanned aviation,
27:18could the connection between
27:20electricity and automobiles
27:22lead to a better performance
27:24than that of gasoline vehicles?
27:26With the high price of fuel
27:28and how harmful it is to the environment,
27:30it's clear that electric cars
27:32are the transport of the future.
27:34But can they be affordable
27:36and sporty at the same time?
27:38Karim Bondar
27:40has traveled to California
27:42to find out
27:44why the last thing on four wheels
27:46actually has three.
27:56The face-to-face definitive.
28:02The emotion of speed.
28:04The rise of adrenaline.
28:08Some of these monsters
28:10use turbochargers
28:12and others burn nitrous oxide.
28:14In any case,
28:16in races like these,
28:18everything revolves around one thing,
28:20combustion.
28:22But maybe not for long.
28:24Today we have a new car
28:26on the runway.
28:32This is the TORC EV Roadster.
28:34Yes, a high-performance sports car,
28:36but it's very far
28:38from being an ordinary car.
28:40It accelerates in an incredible way.
28:42It goes from 0 to 100 in 4 seconds.
28:48The surprise
28:50is that TORC is electric.
28:52But that's not all.
28:54Its technology means a qualitative leap
28:56ahead of its competitors.
28:58A new kind of sports car.
29:00I was determined
29:02to discover its secrets.
29:04Chris Anthony is the executive director
29:06of EPIC, a TORC manufacturer.
29:08I would like to start
29:10with the incredible and aerodynamic chassis
29:12that this car has.
29:14It looks like it's taken from the future.
29:16What is it made of?
29:18The chassis is made of carbon fiber,
29:20the lightest composite material
29:22we can find to reduce its weight to the maximum
29:24and make it as fast as possible.
29:26The TORC has many advanced features
29:28that you would expect from a sports car.
29:30Aerodynamic chassis,
29:32formula suspension,
29:34avant-garde braking system.
29:36But this is the result of the design
29:38of a sports car from a revolutionary perspective.
29:40To begin with,
29:42it only has three wheels.
29:44One thing people notice
29:46in front-wheel drive racing cars
29:48is that every time you turn abruptly,
29:50one of the rear wheels lifts.
29:52We wonder if in a four-wheel car
29:54it doesn't have this traction either.
29:56Why keep that wheel?
29:58Having only three wheels
30:00has two advantages over its rivals.
30:0290 kilos less weight
30:04and a better grip on the road.
30:08In this vehicle,
30:10which is missing the top part,
30:12you'll see that we've put a lot of batteries
30:14in places where you can't see them.
30:16But the most revolutionary technology
30:18that the TORC presents
30:20is its electric center.
30:22Most of the weight of the batteries,
30:24about 360 kilos,
30:26is between the central tunnel
30:28where the cells contain an exceptional nanomaterial
30:30called lithium iron phosphate.
30:32Lithium iron phosphate
30:34stores more charge,
30:36so that each cell
30:38contains more charge density.
30:40Its most important advantage
30:42is that they are 60% lighter
30:44and provide 20 times more energy
30:46than normal car batteries,
30:48all thanks to a specifically designed program.
30:50This is a smart system.
30:52If one cell is heated more than the rest of the cells,
30:54if one of them releases more energy,
30:56this control system performs a correction task.
30:58All of this is reflected
31:00in one aspect,
31:02performance.
31:04The power that comes to the engine
31:06is about 260 kilowatts.
31:08This power reaches the wheels
31:10passing through the differential,
31:12putting about 82 kilograms of torque
31:14on the ground.
31:16That's almost the same power
31:18as a Ferrari Testarossa,
31:20but with more power.
31:22Three wheels and some batteries
31:24with a lot of load
31:26should give this car an advantage
31:28over the others, in theory.
31:30In terms of road grip,
31:32will three wheels work better than four?
31:34I'll check it right now.
31:36Hi, Mike. Hi, how are you?
31:38Very good, more than ready.
31:40Oh, my God!
31:42What a ride!
31:44What I experienced
31:46was what is known as
31:48lateral G-force.
31:50The higher the G-force,
31:52the greater the road grip.
31:54Most high-performance sports cars
31:56barely exceed a G-force.
31:58Mike claims that the TORC
32:00is capable of much more.
32:06Well, it's been incredible
32:08and terrifying.
32:10How far have we gone?
32:12We got 1.17 G-force
32:14on the sides,
32:16more than a new Corvette.
32:18Exactly.
32:20God, now I believe
32:22in the power of electric cars.
32:24Engine fans, get ready.
32:30In the authorized car race
32:32in San Diego,
32:34supercharged cars
32:36compete for about 200 meters.
32:38The TORC is about to make
32:40its debut in the races
32:42and it's starting to stir.
32:44It's the first I've seen in my life.
32:46I've never seen anything like it.
32:48They can do it very well.
32:50And where do you put the purchase?
32:52Shelby, Camaro, Piper.
32:54The TORC has to face
32:56the big gasoline cars.
32:58To have any chance,
33:00it must reach the finish
33:02in less than 12 seconds.
33:04But at the beginning,
33:06it has a problem.
33:08The batteries give so much power
33:10to the wheels in such a short time
33:12that it can't grip the track.
33:14For a long time,
33:16seven laps later,
33:18it beats some of its strongest rivals.
33:28We did very well.
33:30We beat four of the seven
33:32cars that consumed gasoline.
33:34Our best time was nine seconds
33:36in a lap and we went from
33:38zero to 100 in four and a half seconds.
33:40We're losing traction.
33:42But apart from that,
33:44we've done very well.
33:46The TORC will be tripled
33:48from here to 2015.
33:50The price?
33:52$65,000.
33:54The future is electric.
33:58By 2020,
34:00world electric car sales
34:02will exceed 20 million
34:04and the battery market
34:06for lithium-ion batteries
34:08will be valued at $25 billion.
34:12The ability to reduce our dependence
34:14on fossil fuels
34:16leads us to the ultimate connection,
34:18to an atomic energy
34:20without limits that may be possible
34:22thanks to the elusive science
34:24of nuclear fusion.
34:28In 1917,
34:30humanity separated the atoms
34:32to create weapons and energy.
34:34I think it says a lot about our species
34:36that we were able to create
34:38a destructive reaction
34:40in another part of the universe.
34:42What does occur incessantly
34:44in the sun
34:46is nuclear fusion,
34:48which occurs when two atomic nuclei
34:50collide, join together
34:52and create an extraordinary amount
34:54of energy.
34:56Unlike fusion,
34:58fusion provides a safe,
35:00clean and virtually unlimited energy.
35:02The challenge is to create
35:04a fusion power plant
35:06that doesn't require
35:08that fusion occurs in nature.
35:10Several of the world's richest countries
35:12have allied themselves
35:14to carry out the most expensive
35:16scientific experiment ever carried out.
35:18Arati Prasat is in charge
35:20of investigating it.
35:26With a budget of about
35:2813 billion euros,
35:30everything related to the ITER project
35:32in France is something out of the ordinary.
35:34The workers build
35:36a 360,000 ton structure
35:38to test earthquakes
35:40the size of 60 football stadiums.
35:42It is expected that the reactor
35:44produces enough energy
35:46to illuminate
35:4850,000 homes without interruptions.
35:50It seems like
35:52the perfect solution
35:54for the planet
35:56in terms of electricity,
35:58but there is an obstacle.
36:02Fusion could be
36:04the perfect solution
36:06to create energy.
36:08It's safe.
36:10It doesn't produce pollution
36:12in the long run.
36:14It's just a problem.
36:16It's very difficult to produce it.
36:18This great project
36:20is an exceptional bet
36:22because no one knows
36:24for sure if it will work.
36:26Seven countries are participating
36:28and 34 more are lending resources.
36:30But we face the key challenge,
36:32I went to Calan,
36:34the fusion research center
36:36in Oxfordshire,
36:38where the fusion reactor
36:40known as JET is located.
36:42The main challenge is
36:44to keep a hot and unstable fuel,
36:46a hydrogen isotope plasma,
36:48inside the reactor without touching the sides.
36:50If that happens,
36:52it suddenly cools down
36:54and the reaction stops.
36:56This is an exact replica
36:58of the inside of the JET.
37:00It was designed by engineers and technicians.
37:02As you can see,
37:04this chamber is shaped like a dome
37:06and we use it to contain the plasma
37:08at the high temperatures required for fusion.
37:10Then we reproduce the process
37:12that creates the energy inside the sun.
37:14That's what we intend to do here.
37:16So we have very powerful magnets
37:18that surround the chamber
37:20and create a kind of magnetic bottle
37:22that contains a very hot gas
37:24at the necessary temperature
37:26for fusion to occur.
37:28So the idea is that the plasma levitates
37:30and does not touch any of the sides.
37:32Exactly. With the JET,
37:34we have shown that it is a process that works,
37:36that we can create fusion.
37:38And if we can create fusion for a few seconds,
37:40we can also create it
37:42for a much longer period of time.
37:44That requires a stronger magnetic field.
37:46The problem is how huge
37:48the necessary power would have to be
37:50to create a strong enough magnetic field
37:52to keep the plasma in place
37:54for longer.
37:56The JET has been able to generate
37:58up to 70% of the energy
38:00and that was only once.
38:02The results are usually worse
38:04and maintaining a temperature
38:06of 150 million degrees Celsius
38:08requires even more energy.
38:10In the tests that have been carried out,
38:12140,000 volts have been needed
38:14to keep the plasma away
38:16from the walls of the chamber.
38:20On the JET, we have been able
38:22to obtain a little energy
38:24but we had to invest energy
38:26to obtain energy.
38:28What we are hoping for the JET
38:30is that we can obtain energy
38:32with an initial investment
38:34of minimal energy,
38:36perhaps even zero.
38:38Once the reaction takes place,
38:40the intention is that it is self-sustaining.
38:42I would like to know
38:44what makes this progress possible
38:46and find out if this new reactor
38:48could bring us closer to the utopia
38:50of having an unlimited supply
38:52of hydrogen.
38:54In the sea water,
38:56we find an isotope of hydrogen
38:58in great abundance, deuterium.
39:00With less than 4 liters of it,
39:02the same energy could be produced
39:04as with more than 1,000 liters of gasoline.
39:06However, what is even more important
39:08is that it would sustain the reaction.
39:10Any type of impurity in the chamber
39:12would interrupt it
39:14and throw it away.
39:16Currently, the team is testing
39:18how carbon affects the plasma circle.
39:20Tests like this are fundamental
39:22for the design of the reactor,
39:24and I am part of one of them.
39:26We had to heat a quantity of fuel
39:28the size of a grain of sand
39:30at a temperature ten times higher
39:32than that of the center of the sun.
39:34By pressing a single button,
39:36I would create my own star.
39:38But how long would it last?
39:40If you are so kind,
39:42press that button there and say
39:44launch shot.
39:46Launch shot.
39:48Yes, there is my star.
39:54The life of my star was short.
39:56The carbon in the chamber
39:58caused my plasma to lose stability.
40:00With the help of high-speed cameras,
40:02the team analyzes
40:04the reaction obtained.
40:06It lasted less than half a second.
40:10The plasma contains a lot of energy
40:12and tends to move everywhere.
40:14Flashes and distortion occur.
40:16Those filaments that you see in eruption
40:18from the plasma are somewhat similar
40:20to solar eruptions,
40:22so we have to use magnetic fields
40:24to confine it,
40:26to keep it as stable as possible.
40:28Solving these problems is crucial
40:30to be able to materialize
40:32the dream of nuclear fusion,
40:34which is expected to become
40:36reality in 2026,
40:38when it is scheduled to activate the ITER.
40:40I can't help but think,
40:42what if we invested some of the billions
40:44in fusion reactors?
40:46We could make this dream a reality
40:48much sooner.
40:502026 may seem
40:52very far away yet,
40:54but looking at it in perspective,
40:56it will come in a blink of an eye.
40:58As the ITER advances
41:00to be completed,
41:02we no longer wonder
41:04if nuclear fusion will be possible,
41:06but simply when.
41:08Machines programmed to learn,
41:10deposits linked to biology,
41:12fast and fun electric motors,
41:14and limitless energy
41:16based on the power of stars.
41:18These are just some of the hyperconnections
41:20that will define the future
41:22of our planet.
41:24Thank you for your attention.
41:42Transcription by ESO. Translation by —