Incredible Robots - Rise of the Machines

one day in the not-too-distant future robots will travel to the far reaches of the universe all systems activate systems prepare for disembarkation they will be the first to colonize new worlds robots will lead the way in the exploration of deep space after they are robots the future of mankind [Music] robots machines of our nightmares or servants of men in the 1930s film metropolis the robot was an evil character it represented our darkest fears if we're not careful it may indeed be the case that early generations of super intelligent robots will simply eliminate us you know as and as a nuisance basically having better things to do with the space in the matter we're occupying without you what could he do there's no limit to what he could do he could destroy the earth and by the 1950s they had become even more sinister and powerful but over the last few decades our opinions of robots have dramatically changed they've been reinvented as the police force of the future and really good matter guys but can real robots match the exploits of their celluloid Cussler's although it looks really good that robots can on TV you know like lift up a building and throw Captain Kirk across the room in real life robots are fragile weak machines which don't have the energy to do anything more than barely move themselves around while the movies were creating ruthless men of steel real robots were starting their own painful march into the world robots are still basic but over the last few decades they have advanced enormous Li before robots can become the masters of the universe or even the surface of mankind they need to accomplish one important thing they need to move around [Music] locomotion in general of course is very important because a robot that can't transport itself around is a very inefficient sort of thing without mobility a robot is limited only to where it can reach and so the mobility is a function is the crux it is the essential capability for complex achievement getting around isn't easy the simplest method is to use wheels but they can be defeated by uneven terrain or stairs even tracks will eventually fail all sorts of ingenious modes of locomotion have been tried the best solution is to build a robot with legs I think that in a practical sense walking is absolutely critical to the incorporation of robots into everyday life we have built our world four-legged creatures namely ourselves and so any robot that is going to move around in that world in an efficient way in a relatively quick way is going to have to traverse terrain that we have built for ourselves things like stairs and of course since we have legs it makes sense to have the robots headlights also to find out the best way for a robot to move around the scientists look to nature nature basically slithers walks and swims and flies so I think that's the future of robotics this nature has already done a lot of the work and tried out a lot of the possibilities and so it makes sense to take advantage of whatever lessons Nature has already figured out nature is sort of the ultimate engineer you know it's had time to play around with all the different ideas and it's picked the ones that really work the best there have been many attempts to copy nature some successful others less-than-perfect [Music] some robots have gone where even nature didn't dare go the first question that needs to be answered is how many legs should a robot have it's two legs better than four and we say well you know ostriches are really fast they run on two legs but the fact is the fastest animal is a cheetah and it has four legs the first six legs walking insect robots were created in the 1980s at MIT these evolved into Genghis it walks like an insect always keeping enough legs on the ground at any one time so it doesn't fall over Hermes is more advanced the legs have been programmed to move in different patterns so it can walk with different gaits cricket was made from model airplane parts it moves it's six legs in a simple gait using just three motors the robot aerial is based on the crab this mind clearing robot walks sideways it is designed to operate in shallow water and can still walk even if flipped over on its back [Music] [Applause] [Music] designed at carnegie mellon university Ambler is a massive six leg robot a 20 feet high it's the largest walking robot ever built but because it only moves one leg at a time it's pretty slow [Music] the adaptive suspension machine is also modeled on an insect this human-driven robot uses six legs and can walk sideways as well as forward it is designed as a truck that could climb over terrain impossible for a wheeled vehicle one of the most revolutionary six legged robots ever made is owed x1 built in 1983 by Steve Bartlett it was years ahead of its time the robot gets around by moving three legs at a time and maintains its balance by keeping the other three legs firmly on the ground this is known as static balance a static leg system is one that when he stops or for that matter when he's in motion he always has at least three feet on the ground at one time with the center of gravity somewhere within that pattern so he doesn't have to actively balance himself he's always stable and when he comes to a stop it's just a stable isn't free like a stool because she's got at least three feet on the ground arrayed in a pattern and center of gravity is safely for the center of it Oh Dex uses the same technique to climb stairs but six leg robots have a major disadvantage six legs requires six motors which increases the robot's weight and power requirements this is a problem that Nature has already overcome once you're larger than a mouse those extra legs are redundant they're a waste they're more energy and power than you actually need mother nature knew this you'll never find a creature in nature larger than a an ounce with more than four legs poor leg robots are lighter and more energy efficient but they're more unstable with quadrupeds the issue is somewhat more complicated you still need to figure out how to get from point A to point B but you have to do so in a manner that doesn't get your legs tied up or caused an instability to occur the solution is to move one leg at a time keeping the other three on the ground but these robots are painfully slow is it possible for a robot to walk with just two legs I like robots that have two legs and the advantage there is that if you're going to have more than two limbs you can use the other two limbs as arms 2 for me is good because two legs is half of 4 legs so I only need to make half as many legs which is a factor I mean cost goes down if you only have to go two as opposed to four legs and it makes the robots smaller and lighter so there are some engineering advantages to making two legs the major disadvantage is that balance is much more difficult so you are now if you have two legs during a great part of the walking cycle you're really only on one and you have to achieve balance with that and so the robot has to have mechanisms to balance which are different than what you would have with six legs with six legs you can just stop at any point and you are statically balanced with two legs you have to do dynamic balanced dynamic balance was developed by Mark raybert who founded the MIT leg laboratory in the 1980s he started with one leg dro BOTS in our lab we've been concentrating on active balance and to do that initially we built machines that only had one leg a one legged machine has to balance because it doesn't have any opportunity for static support that is you can't spread out the feet in order to keep it balanced passively so we built one legged machines they actually hop instead of walked and we found that the control of these machines wasn't really too difficult the robot had sensors which measured its motion and had a computer brain which use this information to reposition the leg between ops mark Raber then moved on to two legged robots the computer calculated where the legs should move to keep it balanced all Rae Bert's dynamic robots had to keep moving to stay on their feet his dynamic robots performed amazing feats [Music] this robot is spring flamingo it doesn't need to keep hopping or jumping to stop itself from falling over it walks by field sensors in its ankles tell it how to angle its feet to maintain balance spring flamingo was developed by Gill Pratt the present chief of the leg laboratory most robots built to date have really incorporated their vision systems or their navigation systems very tightly into the control loop for locomotion and what we're trying to do with our robot is very similar to the way that a hiker in the woods tends to walk which is that the hiker doesn't look at his or her toes all the time but rather just sort of trust their feet to do the right thing spring flamingo uses another idea borrowed from nature tendons the motors are located above the legs and power is transferred down wire tendons in animals you find that the same thing happens any fast animal you find that the angles of the animal tend to be very graceful and very lightweight and there's really just bone and tendon there and all the muscle is pushed up as far into the leg as far up towards the body of the animal as possible this allows the leg to move as quickly as possible to accelerate very fast and thus less power has to be used in order to move the legs of the robot this robot copies another brilliant idea from nature muscles if you want something that moves backwards and forwards whether it's a finger or an arm or a leg taking motors which are spinning fast and gearing them down and starting and stopping and so on it is very difficult Richard Greenhill has developed robotic muscles they work by pumping compressed air into a rubber sheath surrounded by a wire outer mesh as the rubber sheath expands the mesh contracts and moves the lid so we've now got a muscle that behaves remarkably similarly to a human muscle it contracts about the same amount similar speed similar power and the thing about it is that it's very natural and easy then to connect it to a hand to an elbow to an arm and you get the same movement as a human this robot not only copies human muscles it has two legs and is designed to walk upright good reason for following the human design is it's a design that's been tried and tested it's been through a hole that nature can throw at it for millions of years it is the best design why should we sit down and invent something if we can just borrow from this brilliant and wonderful thing so if we want to build a robot what better of course the problem is can we and that's the challenge it's extremely difficult even though we take walking for granted it's something humans take years to master keeping upright and not falling over is more difficult than it first appears early humanoid robots tried to keep their balance by rocking from side to side or using a counterweight this robot walks more like a human but it still needs its student helpers to stop it from falling over but there is one robot that makes all the others look like beginners this is the Honda robot p3 the most advanced robot in the world it has taken over 20 years to develop at a cost of over 100 million dollars there were numerous prototypes which slowly learn to replicate human walk [Music] the Honda p3 walks by moving its weight from one leg to the other [Music] using sensors in the ankles knees and hips and a sophisticated vision system it can actually walk up and down stairs [Music] this is the latest version of the p3 slightly smaller and lighter it can walk much faster it is so good it looks almost like a man in a robot suit [Music] it can easily walk up and down any steps the tether is only in case it falls but just walking is not enough to be a true humanoid a robot needs hands fingers and thumbs allow the robot to touch grasp and perform complex tasks the first robotic hands were simple crude devices for grabbing and holding [Music] they became more sophisticated as industrial robots took on more and more complex jobs but they were still unable to perform intricate maneuvers the breakthrough came when roboticists were able to copy our own hands the human hands a masterpiece of Engineering it uses tendons to transmit power to the fingers down muscles in the forearms this robotic hand uses the same system the reason tendons are used in in biological systems as well as robot systems is because it allows you to place the actuator at a distance from the place where work is being done by situating the actuators or motors which power the tendons at the base of the arm the fingers can be thin in dextrous the hand is controlled by rope tendons which bent or extend the fingers the hand can grasp and make fine movements just like a human hand and it can do them with precision and without ever tiring the ultimate machine intelligence was how discovery one voyages towards Jupiter controlling the mission is a talking computer known as hell hell you're the brain and central nervous system of the ship does this ever caused you any lack of confidence let me put it this way mr. aimer no 9000 computer has ever made a mistake or distorted information Halla was conceived by arthur c clarke the famous science fiction writer probably before the end of this century we will be able to construct computers or artificial intelligences which can guard on their own and develop lines of thought irrespective of any programming and which may in principle be more intelligent than we are machines as intelligent as how have yet to appear we already have robots that can move around but we need them to become more intelligent mobility is the thing that sets the stage for the evolution of intelligence because it creates novel conditions for an organism to deal with simply by bringing the organism from one place to another in the in the wide and surprising world humans learn by moving and experiencing new stimuli this could also be the key to developing robotic intelligence roboticists argued robots would be able to learn movement as easily as humans in the 1960s when computer controlled robotics started it was believed that interacting with the world would be a fairly minor problem Hans Moravec found that writing a computer program for a robot to navigate across a room was more difficult than first thought it turned out though that moving across a room was far harder than proving geometry theorems after nine years of working his problem I finally in 1979 ended up with a program that could sometimes pick its way across a cluttered room this whole process only worked about three times out of four the the other time out of four it would become confused about where it was or it would fail to see an obstacle and would would would be there run into a silly place or or simply bump into something vision recognition systems require a powerful computer to identify an obstacle and then maneuver around it takes a huge number of calculations and if an object is moved the robot becomes easily the revolution in microchip technology greatly enhanced the way that robots work tasks that once took minutes could be achieved in a fraction of a second these advances have allowed robots like Xavier to move freely knowing where to go and how to get there it can visually identify and avoid obstacles and stop when it has finished its task xavier here I just stopped to say hi but there is more to intelligence than just navigation and obstacle avoidance a robot must be able to think if it is to function in the real for survival a robot must be able to deal with an infinite number of possible situations if a robot brain were programmed with every single possibility it would have a nervous breakdown trying to cope with all the information the alternative is not to program the robot with all eventualities but to program it how to learn as children grow up they learn to perform simple tasks by trial and error as they mature their brains allow them to learn abstract rules and concepts they not only know how to move the chess pieces but where to move them they learn that if they make a bad move they lose the game these experimental robots also learn by trial and error they have the equivalent of 50 brain cells compared to our billions enough to learn not to bump into the walls of their enclosure a sensor tells them when they are getting close to the wall they learn to control their wheels when detecting the wall so that they rarely hit it this humanoid robot cog is also a learning robot but it's far more complex and learns by watching human behavior this is our robot cogs got eyes up here which should respond to motion right now it's looking to see when it something moves and then it tries to reach out to it so when it first title it didn't really know how to look at motion and it didn't really know how to reach for that motion but over time it's what those two things match is a human child Minton learns over the first few weeks of life how to coordinate its eyes with what it sees and then how the coordinated stubby little arms and reach out towards things the single most surprising thing that happened with this robot was one day Cynthia Farrell was playing with the robot trying to demonstrate that the arms were moving and we were videotaping it and afterwards when we looked at that videotape we said oh the robot and Cynthia are taking turns doing something but we our plan wasn't to put turn-taking into the robot for quite a while cog was learning to behave just like a child with his parent but this level of sophistication takes immense computing power and years of programming now some roboticists believe that even this self learning approach will reach a limit and require too much computer power they believe this will stop the robot from evolving into the Machine of our dreams the complexity barrier is one that I've seen happen with a variety of robotics projects and it is essentially an absolute killer it basically starts when a roboticist has a wonderful idea of I want to build a robot and he has a wonderful three week period we actually builds the frame and builds a thing that carries the computer and then it might be as much as two men years of programming just to get the robot to actually do anything at all and when they finally get it do something if they want to make it do something a little bit more oh it walks across the carpet but now can we get it to go up a step all of a sudden now they need twice as much power and twice as much programming the whole trick then is how do we overcome this complexity barrier mark Tilden's solution is a radical new concept his robots don't use digital computers for brains they use simple analog transistors an idea inspired by nature we have to recognize for example that out of 20 million species on this planet now one of them uses a computer for a brain but most of them in fact all of them can handle the chaos and complexity of the real world there are ants out there that only have 20,000 equivalent transistors in their entire body but a digital watch has over half a million transistors on average why is it the mother nature can do so much with so little well that's exactly the sort of things that I'm trying to find out tilden has gone back to basics he constructs his robots from the simplest electronic components with the fewest number of transistors to keep them simple and tough from old electronic circuits and bits of discarded Walkmans he can construct a robot from scratch in under an hour once powered up the robot starts to interact with its world without the aid of a program to digital brain several years ago we found a method by which we could actually get creatures that using just twelve transistors or arrays of twelve and six we all of a sudden we're getting robots that for an equal increase in complexity we were getting an equal increase in ability the breakthrough was the development of the nervous net like an animal passing electrical signals through its nerves the nervous net is a series of signals that pass through transistors in the circuit board to make the robots walk different signals can produce different walking techniques even though the robots are electrically and mechanically simple they can climb over and avoid any obstacles one of major advantages of nervous networks is in fact to the degree of robustness that we've noticed with these devices and they think survive no matter what sort of damage you do to them now compare this against something like a computer brain if you all of a sudden quake a beer and you just dry some brain cells okay you wake up the next morning and you got a bit of a sinking head but you're still working but if you do the same thing to a single transistor in the millions that you find inside a computer then goodbye charlie Tilden's robots will work even if their limbs are damaged in fact they will keep on going to the bitter end Tilden believes that this sort of ruggedness will be crucial if robots are to survive in the real world no one is quite sure where mark Tilden's nervous nets will lead but his robots are certainly unique and he has managed to achieve so much with very limited resources it's exactly how biologics work that is if you can make machines which essentially don't use computers per segment but essentially just use the same nervous tissue or equivalent nervous tissue as mother nature does then the future of robots might be very cheap and very biologic indeed this is a vision of the future the urban warrior series 3000 week on and react robot is capable of operating for months at a time it has bulletproof Kevlar armor and can track its enemy in the dead of night [Music] urban warrior exists only in the imagination but Robart 3 is for real it has been built to respond to urban conflict battlefield scenarios are changing more into urban warfare and the idea here this is a robot you could take into a scenario like that send it into a building that may or may not contain snipers or other enemy agents it would search out the building and leave the humans outside where it's a little safer Robart three is programmed with sophisticated machine intelligence so it can learn how to operate in an environment it has never seen before it's linked to an operator who decides where to search but robot 3 makes all the decisions on how to get there rhobar three has a companion robot that follows it into a building it provides a communications link with the operator when robot three goes deep inside the companion can also detect intruders if they enter a room after robot three is left and robot 3 has another trick up my sleeve the next obvious step is okay URM this thing and you give it the capability to actually do some damage [Music] Robart three is armed with a Gatling gun that can shoot ball bearings or tranquilizer darts we want some type of device on there that can deter or delay a detected intruder without actually performing permanent injury our death ocular four is active there is an object 70 edges robot 3 can identify and shoot a target but the military doesn't want it to be completely independent yet there'll be a human involved there at least initially to make the determination that this is in fact a valid target before any type of lethal action will be authorized robot is still a laboratory experiment but there are active killer robots already on patrol they don't look like the Terminator they are four-wheel drive vehicles that have been converted into lethal robots this robotic rifle is capable of firing 240 rounds a minute with pinpoint accuracy it never tires it never misses this is fire end a mobile tank destroyer it goes out into the bush and waits for days months even years for its prey [Music] when it identifies an enemy tank it fires with deadly precision will Modern Warfare be fought exclusively by robots looking to the near future we are going to have technological warfare we're not very far from it now but essentially humans removed from the war there's a lot to be said for doing that and we're just having one set of robot machines fighting against another set of robot machines I don't see a scenario in my lifetime anyway where it's going to become exclusively machine versus machine it's nice to think that only the robots will fight with the robots and no one will ever get hurt and somehow it'll be just like a rather expensive chess game being fought with robots of Steel I can't help but worry that some of that warfare is going to spill over and that there will always be human casualties if hunter-killer robots are let loose will we be able to control it this very threat prompted science fiction writer Isaac Asimov to create his laws of robotics in the 1950s the first law is as follows a robot may not harm a human being or through inaction allow a human being to come to harm number two a robot must obey orders given it by qualified personnel unless those orders violate rule number one in other words a robot can't be ordered to kill a human being rule number three a robot must protect its own existence after all an expensive reason equipment unless that violates rules one or two but these laws were fiction do they work in the real world in reality they're not applicable even if you look at some of the military machines of today a cruise missile it does not obey Asimov's laws in fact it breaks them so when we look to robots intelligent robots of the future I don't think we can look to fictional laws to save us the robots will be doing things for themselves and that certainly won't be stopping because they might injure a human I wouldn't build any type of autonomous system that I hadn't provided all kinds of safety precautions in there to keep it from doing things that I didn't want it to do I wouldn't focus strictly on the three laws that Asimov put into his literature but you've got to do that kind of thing you're going to have trouble on your hands tobor the most amazing the most fantastic creation of man's mind Oh for tobor can live for no human can breathe in the airless atmosphere of outer space Electronics scientists have designed a practical spaceship atomic power makes space travel possible eating only the most valuable of all secret scientific achievements space conquering giants that man can control like tobor are real space explorers are robots robotic space probes like voyagers 1 & 2 travelled through our solar system out into the far reaches of the universe they may not have arms or legs but they do have fantastic vision systems those types of systems have their place a rover on the surface provides you close-up analysis of the surrounding terrain you can actually put a science instrument up against the rock fury in the sky you can't do that the first robotic system to land on the moon was surveyor 3 in 1963 but it wasn't mobile and could only study what its arms could reach [Music] if you land in one spot in your immobilize the only thing that you get I think that most scientists feel that mobility is absolutely essential NASA used research from different areas of robotics to develop mobile space Rovers Rovers provide mobility we might think of a rover as a mobile instrument platform the rover is providing the mobility with which the scientific tool can look around and feel and see different sights that it can be exposed to the first successful autonomous rover to explore another planet was Sojourner which explored the surface of Mars in 1997 it investigated the area around the lander and sent back detailed pictures of the Martian surface this is its successor rocky 7 like Sojourner it uses wheels legs may be the best system on earth but for Space Exploration they would be too heavy clumsy with the legs you use more energy than with wheels and for space applications you want to use as little energy as possible so not only do we use low energy electronics on board we also want to use low energy forms of mobility wheels our proven technology rocky 7 is extremely stable it has a sophisticated computer brain and it's designed to explore its way around the planet it has never seen before [Music] the Sojourner was basically controlled by the lander so he could not move outside the line of side of the lander which means it couldn't go more than about 30 meters distant from the lander so distance is a major factor rocky 7 has the capability for self location even outside the line of sight of the lander so it uses landmarks it uses Sun sense and it uses various approaches to figure out where it is rocky 7 will have to operate for long periods on its own it's intelligent enough to explore without direct control from earth all Mars rovers have to be autonomous because the amount of time it takes for signal to get from Earth to Mars is too long and so we send it some commands in the morning and then it's on its own all day long and it has to be autonomous enough to be able to accomplish something meaningful after it receives its commands rocky 7 can navigate large distances collect samples and analyze the bites the purpose is to provide a surrogate for the scientists to be able to do experiments to be able to view the terrain send pictures back place instruments send science data back instead of sending an astronaut you send the road despite all its sophistication rocky 7 cannot be expected to explore a whole planet planets vast territories and we don't explore them by just seeing a little piece our traditional view of planetary explorers are that they'll and leave the lander but not very far they're reliant for computing for commands for processing for viewing for position estimation on the lander as a mothership read Whittaker has been working on a robot that can travel much further distances this is nomad nomad is a new class of planetary Explorer which is self-contained self-reliant with the ability to leave the lander and travel far it has the ability to perform long distance long duration traverses majored in hundreds of kilometers not a few meters and to operate over months or years not days or weeks Nomad may be the perfect machine to explore the Martian terrain in the future but it has one main disadvantage sighs because payloads onboard future missions are so limited NASA is looking for much smaller Rovers this is the nano rover tiny robots like this may be the future of space robotics measuring just 12 inches long it's going to the surface of an asteroid that mission will be to find interesting places that we can take images of that we can take scientific data we carry two different spectrometer instruments that will determine the elemental composition and the mineralogical composition of the asteroid the unique wheel supports can swivel around if a nano rover tips over on its mission it will face a more serious problem an asteroid has almost zero gravity the designers have decided to use the low gravity to help the robot move around the asteroid by flicking its legs together it can jump up and bounce over the surface when you go faster than a certain speed in a zero-gravity environment you will take off from the surface and you can plan that trajectory much like you plan trajectories for spacecraft by controlling the speed and understanding how fast you turning the wheels you could predict where you're going to land so literally I mean flying will just hop and we'll use scissors like motion of the struts to hop the vehicle into the sky as much as as little as a meter or two and perhaps as much as 100 meters or more other robots are already at work in space a dexterous robotic arm is an essential tool for astronauts working on the space shuttle it helps them capture and repair satellites like the Hubble Space Telescope the primary advantage of using a robot in space is to avoid having to have a human go out in space and do extra vehicular activity for EDA as its called it's very dangerous operation one estimate done about ten years ago was that there would be a fatality approximately every 1,000 hours of EBA it's much more safe to have a robot doing things in space than a human [Music] NASA scientists are developing even more sophisticated robot arms to work on the International Space Station but the strength of a robotic steel arm can be a hazard [Music] because a robot is made of metal and as strong it can do damage and this presence of potential safety hazard to manned operations to prevent this the arms have been programmed with machine intelligence that teaches them to avoid getting too close to the space station or the astronauts [Music] robot arms and robots in general for space missions will be very important in the future for several reasons one is they they can be stored indefinitely with with very little environmental requirements they don't require food oxygen waste disposal they're expendable should they fail you don't lose human life and that's what's very important they can operate in environments where humans cannot operate because of radiation heat cold the fears of course are the same as with any science mission that something will go wrong as some famous old guy here at JPL once said you're always one transistor away from failure when you reach the forbidden planet you will need a charming character in the robot able to produce on order ten tons of lead or a slinky evening gown always at your service NASA hopes that in the future robots will help us colonize new worlds and build bases across our galaxy yeah I can picture fleets of robots in the sense of colonies doing comprehensive exploration that that is going to happen and also as precursors to human exploration I hope that someday we'll have machines which will terraform Mars take huge sections of the moon and make it basically ready for colonization robots will be able to produce fuels life support materials like water and oxygen and to build habitats and power infrastructure before the time of human presence [Music] in the more distant future spaceships will travel out to the far reaches of the universe insistence activate they'll travel for hundreds of years close to the speed of light they'll carry intelligent autonomous robots system sensors once the ship reaches the atmosphere of a new planet the robots will be activated [Music] systems prepare for disability the robots will construct homes and biospheres for future human expeditions without robots it will be impossible for us to explore the universe robots may even be our first contact with alien civilizations realistically the most likely alien form we're going to see is not a little green man or anything that looks like a human but is a robot is a machine because just as for us it's best to send robots out into space so for someone else and other species elsewhere more than likely they would send robots down here robots have come a long way in the last 50 years they can move around and some are learning to think but how fast are they really evolving I've decided that robots are evolving about 10 million times faster than we did they're currently in their minds roughly as complex as insects but that's already a large step from where they were 50 years ago so by mid-century I think we will have machines that are our equal and in fact our superior in most ways his hands more of it right will robots ever control us [Music] I don't really subscribe to the theory that the robots are going to take over and become smarter than the humans the fact of the matter now is the most sophisticated robots that we have in existence today fall orders of magnitude short in terms of their capability in comparison to even the most inept human being I think in my own mind there is certainly a very real scenario where robots more intelligent than humans will actually be running the earth after all if they're more intelligent than us why would they want to do what we tell them I must say I do read with some amusements this robots taking over idea we're finding it just so bloomin hard to make a robot pick up a car I don't believe that we will ever see a day when all sudden we will have to fear that our toaster basically has plans against us I think robots are our children they will in time grow more capable than we are and will eventually replace us they will inherit the universe from us and I think this is not a bad thing I think they will remember us and they will be able to accomplish far more than we ever could have so I think the universe will be better off for our having done this for their coming to existence [Music] [Music] [Music]

Channel: Naked Science

Views: 542,386

Rating: 4.6169195 out of 5

Keywords: incredible, robots, rise, machines, locomotion, mobility, complex, walking, walk, movement, terrain, stairs, legs, nature, inspiration, build, autonomous, boston, dynamics, engineering, balance, active, humanoid, ai, robotics, eye, science, fiction, intelligent, hal, computer, terminator, robocop, skynet, self, aware, history, microchip, think, function, space, military, predictions, future, brain, experimental, sensor, motion, learning, programming, vision, urban, warfare, isaac, asimov, laws, exploration, interstellar, nasa, forbidden, planet, alien

Id: 8VvYu1fWyxA

Length: 52min 1sec (3121 seconds)

Published: Wed Nov 26 2014