The incredible potential of flexible, soft robots | Giada Gerboni

Prevodilac: Katarina Colic Lektor: Ivana Korom

So, robots.

Roboti, dakle.

Robots can be programmed

Roboti mogu biti programirani tako

to do the same task millions of times with minimal error,

da obavljaju jedan zadatak milion puta uz minimalne greške,

something very difficult for us, right?

a nama je to vrlo teško, zar ne?

And it can be very impressive to watch them at work.

I vrlo je impresivno gledati robote na delu.

Look at them.

Vidite ih samo.

I could watch them for hours.

Mogla bih da ih posmatram satima.

No?

Vi ne?

What is less impressive

Ono manje impresivno

is that if you take these robots out of the factories,

je to da ako ove robote iznesete van fabrika,

where the environments are not perfectly known and measured like here,

gde okruženje nije tako poznato i kontrolisano poput ovog,

to do even a simple task which doesn't require much precision,

i date im jednostavan zadatak koji ne zahteva preciznost

this is what can happen.

ovo se može dogoditi.

I mean, opening a door, you don't require much precision.

Otvaranje vrata, mislim, koliko morate da budete precizni.

(Laughter)

(Smeh)

Or a small error in the measurements,

Ili usled manje greške u merenju,

he misses the valve, and that's it --

nije uspeo da dohvati ventil, i eto -

(Laughter)

(Smeh)

with no way of recovering, most of the time.

obično ode u zaborav.

So why is that?

Ali zašto je to tako?

Well, for many years,

Već mnogo godina

robots have been designed to emphasize speed and precision,

roboti se dizajniraju tako da naglase brzinu i preciznost,

and this translates into a very specific architecture.

a to podrazumeva vrlo specifičnu arhitekturu.

If you take a robot arm,

Uzmimo za primer ruku robota

it's a very well-defined set of rigid links

koja predstavlja vrlo dobro definisan set krutih karika

and motors, what we call actuators,

i motora, koje se nazivaju pokretači,

they move the links about the joints.

i koji pomeraju karike između zglobova.

In this robotic structure,

U ovakvoj strukturi

you have to perfectly measure your environment,

morate precizno da odredite okruženje,

so what is around,

dakle, šta se nalazi oko njega,

and you have to perfectly program every movement

i morate da detaljno isprogramirate svaki pokret

of the robot joints,

zglobova robota

because a small error can generate a very large fault,

jer i najmanja greška može da dovede do ogromne greške,

so you can damage something or you can get your robot damaged

možete da oštetite nešto ili oštetite robota samog

if something is harder.

ako je sklop previše tvrd.

So let's talk about them a moment.

Zato hajde da se pozabavimo time.

And don't think about the brains of these robots

I nemojte da obraćate pažnju na mozak robota

or how carefully we program them,

ili na to kako su isprogramirani,

but rather look at their bodies.

već na njihova tela.

There is obviously something wrong with it,

Očigledno da nešto nije kako treba

because what makes a robot precise and strong

jer ono po čemu je robot precizan i snažan

also makes them ridiculously dangerous and ineffective in the real world,

takođe utiče i na stepen opasnosti i efikasnosti u stvarnom svetu,

because their body cannot deform

jer se njihova tela ne mogu deformisati

or better adjust to the interaction with the real world.

ili bolje prilagoditi interakciji sa stvarnim svetom.

So think about the opposite approach,

Hajde onda da razmislimo o suprotnom pristupu,

being softer than anything else around you.

o većem stepenu mekoće od bilo čega oko nas.

Well, maybe you think that you're not really able to do anything if you're soft,

Verovatno mislite da ništa ne možete da postignete ako ste previše meki,

probably.

zar ne.

Well, nature teaches us the opposite.

Međutim, priroda nas uči drugačije.

For example, at the bottom of the ocean,

Primera radi, na dnu okeana,

under thousands of pounds of hydrostatic pressure,

ispod hiljada paskala hidrostatičkog pritiska,

a completely soft animal

jedna vrlo meka životinja

can move and interact with a much stiffer object than him.

se kreće i interaguje sa objektima koji su daleko tvrđi od nje.

He walks by carrying around this coconut shell

Ona hoda noseći ovu ljusku kokosa

thanks to the flexibility of his tentacles,

zahvaljujući fleksibilnosti svojih pipaka,

which serve as both his feet and hands.

koji služe kao noge i ruke.

And apparently, an octopus can also open a jar.

I očigledno, hobotnica može otvoriti teglu.

It's pretty impressive, right?

Prilično impresivno, zar ne?

But clearly, this is not enabled just by the brain of this animal,

Jasno je da za ovako nešto nije potreban samo mozak životinje,

but also by his body,

već i njeno telo,

and it's a clear example, maybe the clearest example,

a ovo je čist primer i verovatno najjasniji do sada,

of embodied intelligence,

otelotvorene inteligencije

which is a kind of intelligence that all living organisms have.

koju poseduju sva živa bića.

We all have that.

Svi mi to imamo.

Our body, its shape, material and structure,

Naše telo, njegov oblik, materijal i struktura

plays a fundamental role during a physical task,

igraju fudnamentalnu ulogu tokom izvođenja nekog fizičkog zadatka

because we can conform to our environment

jer se mi možemo prilagoditi okruženju,

so we can succeed in a large variety of situations

stoga ga možemo izvšiti sa uspehom u različitim situacijama,

without much planning or calculations ahead.

bez mnogo planiranja i izračunavanja unapred.

So why don't we put some of this embodied intelligence

Zašto onda ne bismo ubacili nešto od te otelotvorene inteligencije

into our robotic machines,

u naše robotske mašine

to release them from relying on excessive work

kako se više ne bismo oslanjali na ekstenzivan rad

on computation and sensing?

koji podrazumeva izračunavanje i očitavanje?

Well, to do that, we can follow the strategy of nature,

Da bismo u tome uspeli, možemo slediti strategiju prirode,

because with evolution, she's done a pretty good job

jer je, tokom evolucije, uradila prilično zadivljujući posao

in designing machines for environment interaction.

dizajniranja mašina za interakciju sa okruženjem.

And it's easy to notice that nature uses soft material frequently

I nije teško primetiti da priroda često koristi meke,

and stiff material sparingly.

a umereno tvrde materijale.

And this is what is done in this new field of robotics,

I to se primenjuje u ovom novom polju robotike

which is called "soft robotics,"

koje se naziva „meka robotika”,

in which the main objective is not to make super-precise machines,

a koja za glavni cilj ima ne izradu super-preciznih mašina,

because we've already got them,

jer ih već imamo,

but to make robots able to face unexpected situations in the real world,

već izradu robota koji su spremni da se suoče sa neočekivanim situacijama

so able to go out there.

u stvarnom svetu i mogu da izađu u taj svet.

And what makes a robot soft is first of all its compliant body,

Ono zbog čega je robot mekan je upravo njegovo fleksibilno telo,

which is made of materials or structures that can undergo very large deformations,

izrađeno od materijala i struktura koji trpe velike deformacije,

dakle nema krutih karika,

so no more rigid links,

and secondly, to move them, we use what we call distributed actuation,

i drugo, kako bismo ih pomerali, koristimo tzv. distribuirano pokretanje

so we have to control continuously the shape of this very deformable body,

i samim tim neprekidno imamo kontrolu nad oblikom ovog vrlo fleksibilnog tela,

which has the effect of having a lot of links and joints,

što za rezultat ima mnogo karika i zglobova,

but we don't have any stiff structure at all.

ali ne i krutu strukturu.

So you can imagine that building a soft robot is a very different process

Stoga vam je jasno da je izrada mekog robota vrlo drugačiji posao

than stiff robotics, where you have links, gears, screws

u odnosu na tvrdu robotiku, gde postoje karike, zupčanici, zavrtnji

that you must combine in a very defined way.

koje morate da kombinujete na vrlo definisan način.

In soft robots, you just build your actuator from scratch

Kod mekih robota, obično je potrebno da samo napravite pokretač

most of the time,

ni iz čega,

but you shape your flexible material

ali oblikujete fleksibilan materijal

to the form that responds to a certain input.

na način koji odgovara zadatim parametrima.

For example, here, you can just deform a structure

Ovde možete deformisati strukturu

doing a fairly complex shape

vrlo kompleksnog oblika,

if you think about doing the same with rigid links and joints,

ako želite da uradite isto sa krutim karikama i zglobovima,

and here, what you use is just one input,

a ovde imamo samo jedan zadati parametar,

such as air pressure.

kao što je pritisak vazduha.

OK, but let's see some cool examples of soft robots.

U redu, hajde da vidimo neke kul meke robote.

Here is a little cute guy developed at Harvard University,

Ovo je mališa kog su razvili na univerzitetu Harvard,

a koji se kreće zahvaljujući talasima pritiska duž njegovog tela,

and he walks thanks to waves of pressure applied along its body,

and thanks to the flexibility, he can also sneak under a low bridge,

i zahvaljujući toj fleksibilnosti, robot može da se kreće ispod niskog mosta,

keep walking,

nastavi da hoda,

and then keep walking a little bit different afterwards.

a zatim nastavi kretanje malo drugačije nakon toga.

And it's a very preliminary prototype,

I ovo je tek preliminarni prototip,

but they also built a more robust version with power on board

ali se takođe radi na robusnijoj verziji sa pokretačkom snagom

that can actually be sent out in the world and face real-world interactions

koja može da se primeni i interaguje sa stvarnim svetom,

like a car passing it over it ...

na primer, sa automobilom koji prolazi pored ili preko njega...

i nastavi sa radom.

and keep working.

Simpatično je.

It's cute.

(Laughter)

(Smeh)

Or a robotic fish, which swims like a real fish does in water

Ili riba robot koja pliva poput prave ribe u vodi

simply because it has a soft tail with distributed actuation

samo zato što ima mek rep sa distribuiranim pokretanjem

using still air pressure.

koji koristi pritisak vazduha u mirovanju.

That was from MIT,

To je odradio MIT

and of course, we have a robotic octopus.

i, naravno, imamo hobotnicu robota.

Ovo je, zapravo, jedan od prvih projekata

This was actually one of the first projects

developed in this new field of soft robots.

razvijenih u ovom novom polju meke robotike.

Here, you see the artificial tentacle,

Ovde vidimo veštački pipak,

but they actually built an entire machine with several tentacles

ali je dizajnirana čitava mašina sa nekoliko pipaka

they could just throw in the water,

koju možete baciti u vodu,

and you see that it can kind of go around and do submarine exploration

i vidite kako može da kruži okolo i istražuje podvodni svet

in a different way than rigid robots would do.

na drugačiji način od krutih robota.

But this is very important for delicate environments, such as coral reefs.

Ovo je vrlo važno za delikatna okruženja poput koralnih grebena.

Let's go back to the ground.

Vratimo se na zemlju.

Here, you see the view

Ovde imamo prikaz

from a growing robot developed by my colleagues in Stanford.

iz ugla rastućeg robota kog su razvile moje kolege sa Stanforda.

You see the camera fixed on top.

Na vrhu se nalazi kamera.

And this robot is particular,

I ovaj robot,

because using air pressure, it grows from the tip,

budući da koristi pritisak vazduha, raste sa vrha,

while the rest of the body stays in firm contact with the environment.

dok drugi deo tela ostaje u kontaktu sa okruženjem.

And this is inspired by plants, not animals,

Inspiracija za poduhvat dolazi od biljaka, ne životinja,

which grows via the material in a similar manner

koje rastu putem materijala na sličan način,

so it can face a pretty large variety of situations.

stoga se može prilagoditi raznim situacijama.

But I'm a biomedical engineer,

S obzirom na to da sam biomedicinski inženjer,

and perhaps the application I like the most

pa primenu koju najviše volim

is in the medical field,

and it's very difficult to imagine a closer interaction with the human body

nalazim u polju medicine,

i vrlo je teško zamisliti bližu interakciju sa ljudskim telom

than actually going inside the body,

od zalaženja u telo samo,

na primer, u cilju izvođenja minimalno invazivne procedure.

for example, to perform a minimally invasive procedure.

And here, robots can be very helpful with the surgeon,

Ovi roboti mogu biti od velike pomoći hirurgu,

because they must enter the body

jer moraju da uđu u telo

using small holes and straight instruments,

pomoću malih rupa i pravih instrumenata,

and these instruments must interact with very delicate structures

a ovi instrumenti moraju interagovati sa vrlo delikatnim strukturama

in a very uncertain environment,

u vrlo neodređenom okruženju,

and this must be done safely.

i vodeći računa o bezbednosti.

Also bringing the camera inside the body,

Pored toga, uvođenje kamere u telo,

so bringing the eyes of the surgeon inside the surgical field

kako bi hirurg imao uvid u oblast koju operiše,

can be very challenging if you use a rigid stick,

može predstavljati izazov ukoliko se koristi krut štap,

like a classic endoscope.

poput klasičnog endoskopa.

With my previous research group in Europe,

Zajedno sa istraživačkom grupom u Evropi

we developed this soft camera robot for surgery,

razvili smo ovog mekog robota za operaciju,

which is very different from a classic endoscope,

koji se umnogome razlikuje od klasičnog endoskopa,

which can move thanks to the flexibility of the module

i koji može da se pomera zahvaljujući fleksibilnom modulu

that can bend in every direction and also elongate.

koji se savija u svim pravcima, a takođe se može izdužiti.

And this was actually used by surgeons to see what they were doing

Hirurzi jesu zapravo koristili ovaj izum kako bi videli šta rade

with other instruments from different points of view,

sa drugim instrumentima sa drugačijim uglom gledanja,

without caring that much about what was touched around.

ne mareći za to šta se dodiruje usput.

And here you see the soft robot in action,

Ovde možete videti mekog robota na delu,

and it just goes inside.

koji samo uđe unutra.

This is a body simulator, not a real human body.

Ovo je simulacija tela, ne pravo ljudsko telo.

It goes around.

Ide u krug.

You have a light, because usually,

Postoji i svetlo jer, obično,

you don't have too many lights inside your body.

delovi vašeg tela nisu osvetljeni.

We hope.

Nadamo se.

(Laughter)

(Smeh)

But sometimes, a surgical procedure can even be done using a single needle,

Ponekad se hirurška intervencija obavi samo jednom iglom,

and in Stanford now, we are working on a very flexible needle,

a trenutno na Stanfordu radimo na vrlo fleksibilnoj igli,

poput vrlo malog mekog robota

kind of a very tiny soft robot

which is mechanically designed to use the interaction with the tissues

koji je mehanički dizajniran za interakciju sa ovim tkivima

and steer around inside a solid organ.

i manevrisanje kroz čvrst organ.

This makes it possible to reach many different targets, such as tumors,

To omogućava pristup mnogim različitim tačkama poput tumora,

deep inside a solid organ

duboko u čvrstom organu

by using one single insertion point.

pomoću jednog uboda iglom.

And you can even steer around the structure that you want to avoid

I time se, čak, može manevrisati unutar strukture koju želite da izbegnete

on the way to the target.

na putu do željene tačke.

So clearly, this is a pretty exciting time for robotics.

Složićemo se da je ovo vrlo uzbudljivo vreme za robotiku.

We have robots that have to deal with soft structures,

Imamo robote dizajnirane za meke strukture,

so this poses new and very challenging questions

pa se postavljaju neka nova, izazovna pitanja

for the robotics community,

zajednici robotike,

and indeed, we are just starting to learn how to control,

i zaista, tek počinjemo da učimo kako da upravljamo,

how to put sensors on these very flexible structures.

stavljamo senzore na ove fleksibilne strukture.

But of course, we are not even close to what nature figured out

Ali naravno, nismo ni blizu toga što je priroda shvatila

in millions of years of evolution.

tokom milion godina evolucije.

But one thing I know for sure:

Ali jedno zasigurno znam:

robots will be softer and safer,

roboti će biti mekši i bezbedniji,

and they will be out there helping people.

i pomagaće ljudima.

Thank you.

Hvala vam.

(Applause)

(Aplauz)