Category
📺
TVTranscript
00:00Life has been evolving on planet Earth for three and a half billion years.
00:20Natural selection has shaped creatures to survive in nearly every habitat on the planet.
00:30And in the last few decades, humans have found that solutions to many of our technology problems are just waiting to be discovered in the vast library of life.
00:50In this program, we see how whiskers can guide a massive truck.
01:00How elephant communication helps to rescue...
01:09...a trapped miner...
01:14...and how a bat makes it possible for a blind person to go mountain biking.
01:26While in this program we know, they're probably going to get a big leap.
01:33In action, humans rely heavily on vision.
01:39In action, humans rely heavily upon vision.
01:48Humans rely heavily on vision.
01:53But other creatures experience their worlds through very different senses.
02:08A rat in a maze.
02:10Love rats or hate them, their senses seem alien to us.
02:14Rats can see and hear and smell.
02:17But their dominant sense is touch.
02:23By whiskers.
02:26Over 50 muscles in the rat's face keep the whiskers moving,
02:30scanning the surroundings.
02:32We call this whisking.
02:36The rat sees the world with its whiskers,
02:39and a large part of its brain is used just to make sense of the information
02:43they generate.
02:47So much so, it creates a model of the world mostly by touch.
02:55For a creature with very poor eyesight, whiskers are even more important.
03:00The shrew.
03:03Like the rat, the shrew is continually whisking,
03:06vibrating its whiskers at 14 times a second.
03:12The shrew is so small, it has a very high metabolism and must eat all the time.
03:17So it has to react quickly when potential food appears.
03:20The shrew can identify prey, even the kind of prey, by a single touch.
03:31It processes the information in a split second.
03:37Anything edible is instantly devoured.
03:41If whiskers are so accurate, can we use them in our own technology?
03:53In robots?
03:55This is Martin Pearson of the Bristol Robotics Lab in the UK.
03:59Whiskers are interesting for roboticists,
04:02as our mobile robots need to sense the world in which they're moving and occupying.
04:10This is the result.
04:12A robot that receives all its information through its whiskers.
04:15Shrewbot.
04:31So one of the reasons we built this robot to model the shrew,
04:34is that the shrew uses its whiskers to hunt its prey.
04:37And in order to catch its prey, which are crickets,
04:39it needs to detect, discriminate and attack in a very short order of time.
04:43In fact, it's been measured down to 30 milliseconds.
04:48The shrew reacts so fast because a distinct region of the brain,
04:52a whisker processor, responds directly to information from the whiskers.
04:58And scientists use a similar idea in shrewbot.
05:02So animals such as shrews and rats and hamsters have the ability to move their whiskers as we've seen.
05:08So we've reproduced this movement on the robot.
05:10So what's guiding his behavior now is interest in the tactile scene.
05:16For example, at the moment he's not touching anything.
05:19As soon as he makes a contact, he'll move his nose to that point in space.
05:23So it's essentially looking for the most interesting thing in the scene.
05:28Shrewbot is building up a picture of his surroundings by touch, just like the rat and the shrew.
05:40He sends signals back to the computer, which visualizes how Shrewbot interprets his world.
05:46The circle on the right is from an overhead time-lapse camera, watching Shrewbot explore the arena.
05:56The circle on the left shows how Shrewbot maps the arena.
05:59The pink blobs are where it's touched objects, slowly building up an internal map of the environment.
06:07From the inner world of Shrewbot...
06:10...to a huge 4x4.
06:14This is TerraMax.
06:23It doesn't look very unusual, but where's the driver? There isn't one.
06:28Like Shrewbot, it finds its way around with whiskers. Invisible whiskers.
06:34These whiskers are made of light, lasers.
06:40Engineer John Beck of Unmanned Systems explains.
06:44It has LIDAR, which is light detection and ranging. Laser beams, basically.
06:4964 beams spinning at 15 times a second.
06:52Lots of data coming back and a very, very good and accurate 3D depiction of its environment.
06:57We've set up some obstacles, so the vehicle is able to detect these obstacles and then navigate itself around them.
07:14It's so accurate, TerraMax can sense and avoid objects as small as traffic bollards.
07:19So, it's sort of scary at first when you see it driving itself, but it's amazing how trusting you become because it's handling itself very well.
07:49A big advantage of unmanned vehicles is that they can enter a dangerous situation where humans can't go.
08:00Or can't see.
08:09Fire. Thick smoke makes vision useless.
08:13But not whiskers.
08:14Firefly.
08:30So, we send in Shrewbot.
08:33It looks like Shrewbot has encountered an object.
08:43It's a little difficult to work out what that is.
08:47It could be a gas sander.
08:52Okay, now he's visited the other corner and it definitely looks like a gas canister now.
09:01Now he's moving away.
09:10The data from Shrewbot allows Martin to create a detailed picture of the room and identify
09:16any dangerous objects.
09:20Okay, made another contact, just moving along there.
09:26That could probably be a beam, maybe, we'll label that as a maybe, it's moving away now.
09:34Okay, now he's successfully mapped that area and we can send in the firefighters.
09:42So we can design whiskers to sense and identify solid objects remotely, but one creature makes
09:48this system look primitive, the seal.
09:57On land, seals don't seem to use their whiskers very much.
10:01They sense their world with sight and hearing.
10:06And out of the water, they seem clumsy and heavy.
10:13But they must come onto the beach to have their pups.
10:21As soon as they can, they get back into the water.
10:28And underwater, their whiskers give them another sense that we're only just beginning to understand.
10:38Seals are really at home in the sea.
10:49All this splashing looks like great fun, but it creates vortices and trails within the water.
11:06Seals can feel this water movement with their whiskers.
11:09They use these trails of swirling water to find each other in the sea.
11:16Seals hunt fish, often in murky water.
11:20Can the seals also track fishtrails?
11:24Scientists at the Rostock Research Center in Germany are finding out what whiskers can really
11:29do.
11:30Sven Wieskotten is working with a seal called Henry.
11:34For Henry, experiments are fun.
11:46And he gets fish as a reward.
11:48Yes.
11:49Well, we want to show that the sealer is able not only to sense also tiny water movements,
11:57but also to follow along the trails of water movements just by means of his whiskers.
12:03But fish can't be trained to swim on demand, so Sven has another solution.
12:09So we used this radio-controlled submarine to make the trail.
12:13The test is for Henry to track the sub underwater with just his whiskers.
12:25He's been trained to wear a blindfold so he can't use his eyes.
12:34Even so that he doesn't follow the sound of the motor, he's trained to wear headphones.
12:42Once the sub is released, it follows a complex path across the tank and then stops.
12:49The headphones are removed, and Henry begins his search.
12:55All Henry has to go on is the wake of the sub, yet he follows this trail, accurately tracing
13:01the sub's path.
13:03So the submarine generates water disturbances over a long time, and the seal is able to
13:11get into touch with these water disturbance by its whiskers and follow the trail then
13:16until he finds the submarine at the end.
13:22And one more.
13:23Massive.
13:24Massive.
13:32Even after 35 seconds, Henry can still sense the trails and traces of the sub, now 40 meters
13:38away.
13:52A seal's whiskers are as sensitive as a primate's hand.
14:07And we're only just beginning to understand what they can do.
14:18Some creatures sense their world with sound, and many can hear sound beyond our senses.
14:24We hear sound through the air, but sound can also travel through solids.
14:31Some creatures are very sensitive to this type of sound, vibrations through the ground.
14:41From elephants to snakes, the desert of Arizona, home to the western diamondback rattlesnake.
14:57Snakes don't hear airborne sounds very well, but that doesn't mean they can't hear at all.
15:05This rattlesnake is about to meet modern technology.
15:11A walkie-talkie and a mobile phone are placed, very carefully, on the metal sheet near the snake.
15:20When the experimenter speaks into the walkie-talkie, the snake doesn't react to the airborne sound.
15:28But when the mobile phone is activated and vibrates, the snake reacts immediately.
15:44Transmitting sound as a vibration through a solid object is something we could adapt for
15:48our own use, one of our oldest professions, mining.
15:58We've been extracting minerals from the ground for thousands of years, but however much we've
16:03improved the technology, it's still dangerous.
16:07To show how we can reduce the hazards, we set up a drastic experiment.
16:12When an explosion seals off a mine, survivors are trapped inside, completely cut off from
16:30the outside world.
16:33A survivor can't be rescued, unless his rescuers know exactly where he is.
16:45A mobile phone doesn't work underground, the signal won't transmit through solid rock.
16:54But in the natural world, one creature can send complex calls over long distances, through
17:00the ground.
17:05The elephant.
17:11Elephants, like us, talk to each other with sound carried through the air.
17:18But new research shows that elephants also have a very different way of talking to each
17:23other.
17:25Kate Evans studies male elephants.
17:28And she's finding that we need to change the way we think about elephant society.
17:32Historically, our understanding of males was that they left their herd independence and pretty
17:42much that was it, they became solitary.
17:48We used to think the males didn't have much to do with each other once they'd left the herd.
17:52What we're slowly unravelling is that male society is actually highly social.
17:59And when they leave their herd, yes, they become independent of the females that they've grown
18:02up with, but they actually join a whole social network of males.
18:07And it's really important for these younger bulls in particular to get to know these older bulls.
18:11And these old bulls, who've been around for 50, 60 years, have a wealth of knowledge about
18:16where food resources are, where water resources are, particularly in times of drought or stress.
18:21The older bulls teach the young males, passing on valuable knowledge and showing them how to behave properly.
18:33But most of the time, male elephants are on their own, which is why we used to think they were solitary.
18:39Now we know how they stay in touch with each other, even when they're kilometers apart.
18:47They use sound, but mostly sound at a very low frequency, lower than we can hear, infrasound.
18:55This sound travels through the ground.
18:58The lower the frequency, the further it travels.
19:09When an elephant calls, some of the sound travels down the body and legs, and into the ground.
19:18So elephants communicate through their feet.
19:21We're still not sure how they do this.
19:24Possibly through sensors in their feet that pick up vibrations from the ground.
19:30The young males and older bulls stay in touch with each other, even though they're a long way apart.
19:39A social network of elephants, held together by infrasound.
19:50So, how does understanding elephant communication help our trapped miner?
20:00Jim Squire and Jay Sullivan from the Virginia Military Institute are developing a system to do this.
20:07They also use extremely low frequency sound, ELF, to send signals through solid rock, seismic signals.
20:17There is no easy way to communicate underground after a mine collapse.
20:21Traditionally, the way that communication lines are established are by hardwired lines.
20:25And these lines are destroyed right after a mine collapse.
20:27Hundreds of tons of rocks wipe it out.
20:30So we thought about ways that nature lets us communicate through long distances.
20:34And clearly, seismic waves are one of them.
20:36Seismic waves travel from one end of the earth to the other.
20:38So we thought, how hard would it be to create a system that creates seismic waves and lets us do just that?
20:43They're going to listen to signals from the mine transmitted in infrasound.
20:49This is our receiver.
20:53And this is going to pick up the seismic signals from down in the mine.
20:58And usually what we're going to do with that is we're going to plant this tail right into the ground so it has a direct link down to the mine.
21:05The tail is a microphone, a very sensitive one.
21:08You can pick up a leaf falling many feet away.
21:11So if the miner has a transmitter, he can use it to send a signal to the surface.
21:20Well, what he's got down in the mine with him is our elf transmitter.
21:25And what he's going to have to do is he's going to have to make sure it's coupled well to the ceiling, to the roof of the mine.
21:31The transmitter must make good contact with the rock above.
21:36It's very important that the man in the mine is going to be able to create a good strong clamping force.
21:48You can't let the clamp just vibrate in the air.
21:52It'll just create acoustic energy.
21:54It won't make a good seismic link to the roof of the mine.
21:58But if he does find a nice area up there and he clamps it well with the wrench,
22:02he should have beautiful transmission from the energy of the transmitter into the roof.
22:07All right, let's start it out.
22:09Okay, what gain do you want down here?
22:11Um, let's go with 100.
22:13Okay.
22:14Transmitters could be placed in safe rooms along the mine.
22:21Each would send a coded signal giving the position, the number of survivors, and the air quality.
22:27As the computer analyzes the data, Jim and Jay can pinpoint the source of the signal.
22:36We don't know exactly how deep they are, but it won't make a difference, will it?
22:41No, no, we're going to get the answer within 10 seconds in this terrain.
22:46Yep, and there we go, right there, sector five.
22:49Good air.
22:50Knowing exactly which sector of the mine the signal comes from, they can direct a rescue mission.
22:58The potential for improving safety in mining is tremendous.
23:03But if this is similar to how elephants communicate, could we use it to talk to elephants?
23:16To find out, we took Jim and Jay to Botswana.
23:28They're meeting up with Kate to try something completely new.
23:35They're going to transmit extra low frequency sound through the ground to see what the local elephants make of it.
23:47As in the mine, the equipment needs a good contact with solid rock.
23:52And so we hope that the elephants will come in and be interested.
24:03Basically, we're trying to emulate the way that the corporalizations come from the elephant into seismic signals inside the ground.
24:12And to do that, we really need as much as possible to make this thing like the elephant's leg and the elephant's weight.
24:19And on this system, the elf itself is this big circular disc.
24:23And it's like having a really large elephant foot.
24:26So we really need to have a lot of force, okay, to get that coupling.
24:34Yes, that's good. It's not binding at all.
24:37They set up the equipment to try a test signal.
24:46Click on the signal.
24:52I feel the ground here.
24:54Yeah?
24:55That's it. No, that's at low. That's at about 10 watts.
24:58Okay.
24:59So we're going to do, what, 125?
25:00Yeah, check it up.
25:01Phew, that's very exciting.
25:03I've just caught an elephant.
25:06Hello.
25:12Sandbags make the artificial foot really heavy, as if it was carrying an elephant's weight.
25:21All right, so...
25:22Great.
25:23If they're sending out signals to elephants, what are they saying to them?
25:28Kate has previously recorded calls from female elephants that are ready for mating.
25:34Well, we've constructed a system here that takes signals that Kate's recorded from elephants about to enter asteris.
25:42These calls are amplified by these large amps, up to about 150 watts, and sent through our seismic transducer.
25:51And we're pretty confident they can go out to at least 200 meters.
25:54We're hoping out to well over two kilometers.
25:58We've never tried it before, so we're about to find out.
26:01Kate, we'd like you to, uh, have a go at it, starting it off.
26:07Great.
26:08Well, let's, uh, move off to a safe distance and see what happens.
26:12Makes sense.
26:16The moment of truth.
26:17Kate sends the signal.
26:31For an elephant, these calls may carry vital information.
26:37A female is often receptive to mating for only five days every four years.
26:45When she's ready, a male has to find her quickly.
26:48So if any signal from the ELF system is going to attract the attention of an elephant,
26:54it should be this one.
27:01Elephants, and their males.
27:09They can hear her, but she's nowhere to be seen.
27:33One hopeful male is clearly ready and willing to mate with the phantom female.
27:49But elephants are not stupid.
27:51And he quickly realizes he's been duped.
27:56That was an interesting, um, reaction.
27:58You've never listened.
27:59He picked up some information.
28:01How much?
28:02We'll never know.
28:03Um, but, uh, it was a female call there.
28:06So he may have been reacting to, to the possibility of a herd in the area.
28:14Using seismic sound to locate trapped miners is a neat trick.
28:19But clearly, elephants can transmit and receive much more sophisticated signals.
28:24From very low frequency sound to sound too high for us to hear.
28:34Ultra sound.
28:39Many bats find their way in the dark by echolocation.
28:52Small bats, micro bats, shout at a very high frequency in ultrasound, then listen for the echo.
29:00Their calls are very loud and very complex.
29:05So the returning echo carries a lot of information.
29:08This is a much bigger bat, an Egyptian fruit bat.
29:18It roosts deep in caves where it's pitch black, but it uses a different system to echolocate.
29:25It simply clicks its tongue and listens for the echo.
29:35We used to think this was a really crude system, just enough to stop the bat from bumping into the cave walls.
29:41But new research shows how sophisticated these bats really are.
29:53Dean Waters of the University of Leeds studies how these bats find their way in the darkness.
30:00He listens to them with a bat detector, which converts the bats' ultrasound clicks into ones we can hear.
30:08And he found that the bats make two clicks every time they take a reading.
30:15So what we're listening to on this bat detector are the echolocation calls these fruit bats are producing.
30:20They're producing double clicks by clicking their tongue this way and the other way.
30:25This is different to the way that the microbats produce echolocation calls using their larynx.
30:29We think that although it sounds like a very primitive system, this is actually quite a sophisticated system.
30:33It's much more like the system that echolocating dolphins use.
30:37So we're really interested in investigating this.
30:46Dean is creating an obstacle course to test the limits of this system.
30:51He's hanging an array of wires from the cave roof.
30:56A small bell on the end of each line rings if the bat brushes against the wire.
31:01The cave is now in complete darkness. So we use infrared cameras to see what happens.
31:18Well, this experiment is to try and work out how good these fruit bats' echolocation actually is.
31:27So we have some strings here that we've put up around the cave.
31:30We know that these bats can echolocate within a cave and they can find the cave walls.
31:35The big question is, how good is their echolocation? Can they actually detect something as small as, say, a one centimeter wide rope hanging from the ceiling?
31:44And so he just gently touched it a little bit, but he's gone all the way through.
31:49You could see that. He clearly knew that that string was there.
31:52These bats have a wingspan of over half a meter and the wires are 40 centimeters apart.
32:03So to get through, the bat has to fold its wings.
32:08He's brought his wings up. He knows that the strings are there.
32:12He's dropped through the strings and then he's away on the other side carrying on flying.
32:15The bats aren't just avoiding the wires, they can navigate through them.
32:26People have always thought this type of echolocation was very primitive.
32:30But these strings show that these bats are able to detect something which is very small.
32:35So it's a very sophisticated system, much better than we previously thought.
32:39We can't hear the sounds the bat makes, but we can turn these sounds into images.
32:45We use an acoustic camera.
32:52Microphones are arranged in a circle around a normal camera.
32:58The directional microphones pinpoint the source of the sound and its frequency is given a color code.
33:05We can superimpose the image of the sound over the image of the scene.
33:10The flashing shows the bat clicking as it flies through the cave.
33:16And by slowing the image down, we can see its double click.
33:21Dean's experiments and an acoustic camera show that even a simple double click generates complex and detailed information in the echoes.
33:31This gave engineers an idea.
33:39An obstacle course on a high roof.
33:42And a blind volunteer with a white cane.
33:46A lethal combination?
33:47Brian Hoyle of Leeds University has designed a new sort of cane based on the Egyptian fruit bat.
33:58The ultra cane.
33:59The transducer at the front of the ultra cane emits an extremely intense beam of sound forwards.
34:06So that when the user is using the ultra cane, reflections come from the end of the beam and they're received back at the ultra cane.
34:13And the ultra cane then processes that information and delivers the information to the user through these vibrating tactile buttons.
34:20And that tells the user that objects are in his or her path.
34:26The user detects obstacles by feeling the handle of the cane vibrate.
34:31And it's surprisingly easy to use.
34:38Volunteer Bill Gulliver has been blind from birth.
34:42So the ultra cane could change his life.
34:44The advantage of the ultra cane over the conventional long cane is that one can keep away from obstructions.
34:57And they help one to be much more free in one's mobility.
35:05The acoustic camera shows how similar the ultra cane is to the bat signals.
35:09The camera detects the signal sent by the cane and sends it to the laptop.
35:15We can see the clicks given off by the ultra cane as it scans the surroundings.
35:24Because the ultra cane is so easy and intuitive to use, we took this technology even further.
35:30A mountain bike track.
35:35It needs fast reactions to the twists and turns and to the obstacles along the side.
35:53Surely this would be impossible for a blind cyclist.
35:56Daniel Smith lost most of his sight recently due to a very rare genetic disorder.
36:09Yet he's going to attempt to ride the track.
36:13He'll rely on a very special bike.
36:16We designed this bike using the same idea as the ultra cane.
36:19Each handlebar has a sensor to receive echoes from an ultrasound beam which vibrates the handles.
36:29Using a similar clicking system to the fruit bat, can Daniel ride the bike round the track?
36:41You really need to keep a hundred percent level of concentration while you're on the bike.
36:46It's really, really sensitive.
36:49And although they're your eyes there, one force move sort of will lead you going off the track.
36:55So I think it's very, very intense while you're on it.
36:58But very enjoyable at the same time.
37:00It's good to get back on the bike again really, it's a good feeling.
37:02The thing I instantly lost when I lost my sight was sport.
37:12It's not available to me anymore.
37:14And going on a bike, you know, to be able to do that on a track like this, certainly a great feeling really.
37:19It's something I haven't experienced since I lost my sight.
37:24He's only responding to the signals on the bike's handlebars.
37:28Yet uses them to avoid obstacles and follow the track.
37:35The technology works really, because I've just navigated the whole track by myself.
37:40So I'm very pleased, yeah.
37:49Like bats, we're learning to see the world differently. With sound.
38:02In the natural world, many creatures use their senses to find each other.
38:06To gather in massive swarms.
38:09Endless numbers of individuals, all pursuing a common goal.
38:13Famous for their plagues, locusts exist in swarms of millions.
38:29Butterflies gathering clusters on patches of soil rich in salt.
38:38But one butterfly uses its senses to do something that seems impossible.
38:43This is a monarch.
38:51Every autumn, monarch butterflies in North America set out on a journey south.
38:57To Mexico.
38:59Some from as far away as Canada.
39:02A journey of more than 3,000 kilometers.
39:05Their sense of direction is astonishing.
39:07All the butterflies east of the Rocky Mountains navigate to a few tiny groves of trees in the mountains of central Mexico.
39:20They gather in their tens of millions to spend the winter.
39:23So how do butterflies navigate so accurately?
39:24We found recently that, like human travelers in the past, they navigate by the sun.
39:41But during the day, the sun moves across the sky, so the butterflies need an internal clock to tell the time of the day and compensate for the sun's movement.
39:57They always know at which angle to the sun they must fly to keep heading south.
40:02But monarchs are even more sophisticated than this.
40:12They can see polarized light.
40:15Polarized light creates patterns in the sky, centered on the position of the sun.
40:20If the butterfly flies into bad weather, it can still navigate.
40:34It just needs a small patch of blue sky.
40:38Seeing some of the polarized light pattern, it can work out where the sun should be and keeps flying south.
40:45Eventually, this navigation system brings most of North America's monarchs together on a few sheltered mountain slopes.
40:54All done with a processor the size of a grain of sugar.
41:02These monarch wintering sites are one of nature's greatest spectacles.
41:06But they also represent a human dream.
41:09Can we design and build swarms of tiny robots that can fly, find their way, and coordinate themselves like real insects?
41:23The ultimate swarm, honeybees.
41:26And at the Wyss Institute in Harvard, scientists are trying to make this dream a reality by combining research from many areas to build a swarm of robo-bees.
41:41Don Ingber explains.
41:43The robo-bees, the idea there would be to have a collection of very small robots that can fly and carry out functions.
41:51The initial idea was pollination to replace the dying bee population, which is, in certain places, a real problem right now.
42:02The first problem is to get a robot to fly like an insect.
42:09Insect wings work in a very different way from birds or planes.
42:13Unlike birds or bats, there are no muscles in the insect wing.
42:19All the complex bending of the wing needed for stable flight and thrust comes from control mechanisms at the wing base,
42:27or is built into the structure of the wing itself.
42:34We're only just starting to understand this well enough to build an artificial insect wing.
42:39This structure mimics the bending and twisting of a real insect wing.
42:48And it generates the complex patterns of vortices over the wing that keep real insects in the air.
42:59Tiny, lightweight motors flap the wings at high frequencies.
43:03The results are a bit unpredictable, but success isn't far away.
43:11Flight is just stage one of the robo-bee project.
43:17Real insects must also navigate to find their way.
43:21And, like monarch butterflies, most use polarized light.
43:26So does this human system.
43:27A compass that can visualize polarized light using a special camera, which can be used to navigate.
43:45We're making progress towards a robo-bee swarm in both the flight and navigation systems.
43:51But what about creating the swarm itself?
43:55How do big swarms function in nature?
44:02A spectacular example.
44:05A winter gathering of starlings.
44:08Millions of birds gather over roosting sites,
44:11and just before settling for the night, fly together in a magnificent display.
44:15The whole flock seems to act as one, as if they read each other's minds.
44:35The truth is less supernatural, but no less amazing.
44:39Using high-speed film and computer programs, scientists found that starlings only react to the seven nearest birds.
44:51All the complex aerobatics emerge from a few simple rules.
44:57The key to this system, very fast sensory reactions.
45:07But studying this in detail is not easy with starlings.
45:12So scientists used a creature that's easier to work with in the lab.
45:18Another famous swarming creature, the locust.
45:21Even in a dense swarm, locusts don't seem to crash into each other.
45:28Essential for a robotic swarm.
45:31To find out how locusts do this, researchers mount a locust on a wire support in front of a computer screen.
45:38On the screen, a black dot rapidly grows in size, which the locust sees as an object on a collision course.
45:46At normal speed, it doesn't seem to react.
45:51But slowed down, the locust briefly interrupts its wing beat, enough to change its flight path and avoid a crash.
45:58This discovery has an unexpected application.
46:07Engineers are designing systems based on how locusts avoid crashes that automatically stops a car coming too close to the vehicle in front.
46:19Even if the driver is distracted.
46:31Collision avoidance will be needed by robo-bees.
46:35But they must also communicate with each other in more complex ways if they're to become a swarm.
46:40Software for such cooperation is being tested at the Wyss Institute using these helicopter robots.
46:56Kartik Duntu has programmed the robots to search for flowers, just like bees.
47:02The cheapest way to test swarm communications is with virtual robots inside a computer.
47:17But here at the Wyss Institute, this is a hybrid system.
47:21Most of the swarm is in the computer, but a few are real so the scientists can test them in the real world.
47:28In our simulation environment, we can attach sensors of various kinds, throw in algorithms of different kinds and quickly test them out.
47:36But once we test them, it's also important to try them out on the hardware because the hardware introduces various practical issues.
47:45The helicopters explore their environment looking for flowers.
47:49The computer builds up a picture of what they find, and if it's missing information from a certain area,
47:54it sends out a robot to survey it.
47:58If a helicopter fails, the computer sends out another to its place.
48:03But unlike a real bee swarm, all this behavior is being controlled by a central computer.
48:12In the Bristol Robotics Lab, Alan Winfield is interested in autonomous robots.
48:17These are robots that are inspired, bio-inspired, from the kind of swarm behavior that we see in nature.
48:27Lots of animals exhibit swarm intelligence, but it's most prominent in social insects, ant colonies.
48:35So the important thing about swarm intelligence is that there's no brain ant in the ant colony directing the actions of each of the individual ants.
48:50All this behavior seems to be organized, but it emerges from individual ants following simple rules.
48:57So in swarm robotics, we're trying to figure out ways of making robots that self-organize in the same way.
49:10We use these EPUC robots to do experiments in swarm intelligence.
49:15If I show you the tail lights on the robots, the robots are attracted to the tail lights and will try and follow those tail lights.
49:28So we see emergent convoying here of these simple robots.
49:36And this approach has been scaled up to a real swarm at the Wyss Institute at Harvard.
49:41We just licensed a swarm of robots called Kilobots as a toy that's an educational toy.
49:49But you can take identical robots and program them to do things collective, like all follow the leader or create a pattern.
49:58The Kilobots are only a few centimeters across and move by vibrating their three legs.
50:04Changing the pattern of vibration makes the robots move in any direction or turn in a circle.
50:15They communicate using infrared light and can be programmed to follow each other and stay together as a swarm.
50:29The Kilobots are a cheap and practical way for engineers to test out their ideas about robotic swarms.
50:39They bring us a step closer to our goal of building a swarm of robo-bees.
50:43Humans work collectively, insects build collectively, bees pollinate collectively, and we're just beginning to understand some of the rules that nature has leveraged to make that possible.
50:58If you get the essential design features down, you get nature's amazing functionality out.
51:07All of this has yet to be put together.
51:10The tiny mechanical bee, the ability to navigate, and a way to communicate.
51:14But this demonstration of co-operating, flying robots shows the future isn't far away.
51:22It's been far away.
51:23It's been far away.
51:24It's been far away.
51:25The all of this has yet to follow you.
51:26I would see.
51:27Good bye.
51:28Everything matters in my life.
51:29I'm as loved.
51:30I think that this is a kind of theory.
51:31Terrific casino guy where people see hours saying,
51:50Being able to sense the environment is one of the defining features of all living creatures.
52:19So, after three and a half billion years of evolution, there's much more to inspire us in the vast library of life.
52:35In the next program, we explore how animals survive in extreme conditions, and how that
52:41can help us design new solutions to the problems of human survival.
52:53So, after three and a half billion years of evolution, there's much more to do that.
53:01So, after three and a half billion years of evolution, there's much more to do that.