Chris Urmson: How a driverless car sees the road

Prevodilac: Aleksandar Korom Lektor: Ivana Korom

So in 1885, Karl Benz invented the automobile.

Godine 1885. Karl Benc izumeo je automobil.

Later that year, he took it out for the first public test drive,

Kasnije, u toku godine, izveo ga je na prvu javnu test vožnju

and -- true story -- crashed into a wall.

i, ovo se zaista dogodilo, udario ga u zid.

For the last 130 years,

U poslednjih 130 godina

we've been working around that least reliable part of the car, the driver.

radili smo na tom, najmanje pouzdanom, delu automobila - vozaču.

We've made the car stronger.

Ojačali smo sam automobil.

We've added seat belts, we've added air bags,

Dodali smo pojaseve za sedišta, vazdušne jastuke,

and in the last decade, we've actually started trying to make the car smarter

i, u poslednjoj deceniji, počeli smo čak i da činimo auto pametnijim

to fix that bug, the driver.

kako bismo ispravili tu grešku - vozača.

Now, today I'm going to talk to you a little bit about the difference

Danas ću govoriti malo o razlici

between patching around the problem with driver assistance systems

između rešavanja problema sistema asistencije vozaču

and actually having fully self-driving cars

i zapravo, potpuno samoupravljajućih automobila

and what they can do for the world.

i šta oni mogu da učine za svet.

I'm also going to talk to you a little bit about our car

Takođe ću govoriti ponešto i o našem automobilu

and allow you to see how it sees the world and how it reacts and what it does,

i prikazati vam kako on vidi svet, kako reaguje i šta radi,

but first I'm going to talk a little bit about the problem.

ali prvo moram da kažem nešto i o jednom problemu.

And it's a big problem:

A problem je velik:

1.2 million people are killed on the world's roads every year.

svake godine, na putevima širom sveta, 1,2 miliona ljudi pogine.

In America alone, 33,000 people are killed each year.

Samo u Americi svake godine pogine 33 000 ljudi.

To put that in perspective,

Stavimo to u ovakvu perspektivu:

that's the same as a 737 falling out of the sky every working day.

to je kao da se svakog radnog dana 737 sruši iz vazduha.

It's kind of unbelievable.

To je neverovatno.

Cars are sold to us like this,

Automobili nam se prodaju na ovakav način

but really, this is what driving's like.

ali, zapravo, ovo predstavlja vožnju, zar ne?

Right? It's not sunny, it's rainy,

Nema sunca nego kiše

and you want to do anything other than drive.

i radije biste radili bilo šta drugo, osim vožnje.

And the reason why is this:

A razlog tome je sledeći:

Traffic is getting worse.

saobraćaj postaje sve gori.

In America, between 1990 and 2010,

U Americi, u periodu između 1990. i 2010,

the vehicle miles traveled increased by 38 percent.

milje koje su vozila prelazila

se uvećao za 38%.

We grew by six percent of roads,

Procenat naših puteva se uvećao za 6%.

so it's not in your brains.

To nije samo u našim glavama,

Traffic really is substantially worse than it was not very long ago.

saobraćaj se zaista znatno pogoršao u skorije vreme.

And all of this has a very human cost.

I sve to ima veliku cenu po čoveka.

So if you take the average commute time in America, which is about 50 minutes,

Ako uzmete prosečno vreme transporta, u Americi, koje iznosi oko 50 minuta

you multiply that by the 120 million workers we have,

i pomnožite ga sa 120 miliona radnika koje imamo,

that turns out to be about six billion minutes

dobijete negde oko šest milijardi minuta

wasted in commuting every day.

protraćenih tokom transporta svakog dana.

Now, that's a big number, so let's put it in perspective.

To je velika cifra pa sagledajmo to na sledeći način:

You take that six billion minutes

uzmite tih šest milijardi minuta

and you divide it by the average life expectancy of a person,

i podelite ih sa prosečnim životnim vekom jedne osobe

that turns out to be 162 lifetimes

i dobićete 162 života

spent every day, wasted,

uzaludno utrošenih svakog dana,

just getting from A to B.

prelazeći samo od tačke A do B.

It's unbelievable.

To je neverovatno.

And then, there are those of us who don't have the privilege

Zatim, imamo i one koji nemaju tu privilegiju

of sitting in traffic.

da sede u saobraćaju.

So this is Steve.

Ovo je Stiv.

He's an incredibly capable guy,

On je neverovatno sposoban muškarac

but he just happens to be blind,

ali je, pri tom, i slep.

and that means instead of a 30-minute drive to work in the morning,

A to znači da umesto tridesetominutne jutarnje vožnje do posla,

it's a two-hour ordeal of piecing together bits of public transit

to je iskušenje od dva sata spajajući delove javnog prevoza

or asking friends and family for a ride.

ili moljakanje prijatelja i porodice za prevoz.

He doesn't have that same freedom that you and I have to get around.

On nema tu istu slobodu kretanja koju vi i ja imamo.

We should do something about that.

Trebalo bi da uradimo nešto po tom pitanju.

Now, conventional wisdom would say

Konvencionalno razmišljanje bi bilo

that we'll just take these driver assistance systems

da uzmemo te sisteme asistencije vozača,

and we'll kind of push them and incrementally improve them,

da ih guramo, postepeno unapređujemo i da će se oni vremenom

and over time, they'll turn into self-driving cars.

pretvoriti u samoupravljajuće automobile.

Well, I'm here to tell you that's like me saying

Međutim, to je kao kada bih vam ja sada rekao da

that if I work really hard at jumping, one day I'll be able to fly.

ako bih se mnogo trudio da skačem, da bih jednog dana mogao i da poletim.

We actually need to do something a little different.

Morali bismo ipak da uradimo nešto drugačije.

And so I'm going to talk to you about three different ways

Zato ću vam pričati o tri glavne razlike

that self-driving systems are different than driver assistance systems.

između samoupravljajućih automobila i sistema asistencije vozaču.

And I'm going to start with some of our own experience.

I počeću sa nekim od naših iskustava.

So back in 2013,

Godine 2013.

we had the first test of a self-driving car

imali smo prvi test samoupravljajućih automobila

where we let regular people use it.

i dozvolili običnim ljudima da ga koriste.

Well, almost regular -- they were 100 Googlers,

Dobro, delimično običnim - to je bilo 100 Guglovih radnika

but they weren't working on the project.

međutim, oni nisu radili na projektu.

And we gave them the car and we allowed them to use it in their daily lives.

Dali smo im taj automobil i dozvolili im da ga koriste u svakodnevnom životu.

But unlike a real self-driving car, this one had a big asterisk with it:

U odnosu na druge samoupravljajuće automobile

ovaj je imao jednu veliku začkoljicu:

They had to pay attention,

morali su da vode računa

because this was an experimental vehicle.

jer je ovo bilo eksperimentalno vozilo.

We tested it a lot, but it could still fail.

Mi smo ga dosta testirali, ali je i dalje moglo da se pokvari.

And so we gave them two hours of training,

Dva sata smo ih obučavali,

we put them in the car, we let them use it,

stavili smo ih u auto, dali im da ga koriste,

and what we heard back was something awesome,

i šta smo naknadno čuli je bilo fenomenalno

as someone trying to bring a product into the world.

za nekoga ko pokušava da uvede proizvod u svet.

Every one of them told us they loved it.

Svi ponaosob su rekli da im se svideo.

In fact, we had a Porsche driver who came in and told us on the first day,

Zapravo, imali smo jednog vozača Poršea koji nam je prvog dana rekao:

"This is completely stupid. What are we thinking?"

"Ovo je potpuno glupo. O čemu uopšte razmišljamo?"

But at the end of it, he said, "Not only should I have it,

Ali je, na kraju, rekao: "Ne samo da je meni potreban,

everyone else should have it, because people are terrible drivers."

potreban je svima, jer su ljudi očajni vozači."

So this was music to our ears,

To je bila muzika za naše uši.

but then we started to look at what the people inside the car were doing,

A onda smo počeli da gledamo šta ljudi unutar automobila rade

and this was eye-opening.

i to nam je otvorilo oči.

Now, my favorite story is this gentleman

A meni je omiljena priča jednog gospodina

who looks down at his phone and realizes the battery is low,

koji gleda u svoj telefon i shvati da mu se baterija istrošila.

so he turns around like this in the car and digs around in his backpack,

On se ovako okrenuo, u autu, čeprkao po svom rancu,

pulls out his laptop,

izvukao svoj laptop,

puts it on the seat,

stavio ga na sedište,

goes in the back again,

opet se okrenuo nazad,

digs around, pulls out the charging cable for his phone,

opet čeprkao i izvukao kabl za punjenje za svoj telefon,

futzes around, puts it into the laptop, puts it on the phone.

vrpoljio se, uključio kabl u laptop i u telefon.

Sure enough, the phone is charging.

Zasigurno, telefon se puni

All the time he's been doing 65 miles per hour down the freeway.

a svo to vreme on se vozio brzinom od 100 km na sat, po autoputu.

Right? Unbelievable.

Zaista neverovatno.

So we thought about this and we said, it's kind of obvious, right?

Razmišljali smo o ovome i rekli smo da je to nekako očigledno.

The better the technology gets,

Što više tehnologija napreduje

the less reliable the driver is going to get.

vozač postaje sve manje pouzdan.

So by just making the cars incrementally smarter,

Samim tim što automobil postepeno postaje pametan

we're probably not going to see the wins we really need.

verovatno više nećemo videti pobede koje smo nekada želeli.

Let me talk about something a little technical for a moment here.

Dozvolite mi da vam se obratim tehničkim jezikom, na trenutak.

So we're looking at this graph, and along the bottom

Gledajući ovaj grafik, na dnu primećujemo

is how often does the car apply the brakes when it shouldn't.

koliko često automobil primenjuje kočnice kada ne bi trebalo.

You can ignore most of that axis,

Ignorišite veći deo ove ose

because if you're driving around town, and the car starts stopping randomly,

jer ako se vozite gradom i automobil iz nekog razloga stane,

you're never going to buy that car.

vi nikada nećete kupiti taj auto.

And the vertical axis is how often the car is going to apply the brakes

Vertikalna osa prikazuje koliko često će auto primeniti kočnice,

when it's supposed to to help you avoid an accident.

kada je to potrebno, da bi vam pomogao da izbegnete nesreću.

Now, if we look at the bottom left corner here,

Ako sada pogledamo u ovaj donji levi ugao

this is your classic car.

- to je vaš klasičan automobil.

It doesn't apply the brakes for you, it doesn't do anything goofy,

On ne primenjuje kočnice umesto vas, ne radi ništa šašavo

but it also doesn't get you out of an accident.

ali isto tako vas i ne čuva od nesreće.

Now, if we want to bring a driver assistance system into a car,

Ako sada želimo u auto da uvedemo sistem asistencije vozaču

say with collision mitigation braking,

recimo za kočenje za ublažavanje sudara

we're going to put some package of technology on there,

ubacićemo neki paket tehnologije tu

and that's this curve, and it's going to have some operating properties,

i to je ova kriva ovde i sistem će imati neka upravljačka svojstva

but it's never going to avoid all of the accidents,

ali nikada neće izbeći sve nesreće

because it doesn't have that capability.

jer nema takve sposobnosti.

But we'll pick some place along the curve here,

Ali izabraćemo neku poziciju na krivi ovde

and maybe it avoids half of accidents that the human driver misses,

pa možda izbegava polovinu nesreća koja ljudima vozačima promakne

and that's amazing, right?

neverovatno, zar ne?

We just reduced accidents on our roads by a factor of two.

Upravo smo duplo smanjili nesreće na našim putevima.

There are now 17,000 less people dying every year in America.

Sada 17 000 ljudi manje pogine u Americi svake godine.

But if we want a self-driving car,

Ali ako želimo samoupravljajući automobil,

we need a technology curve that looks like this.

potrebna nam je kriva koja izgleda ovako.

We're going to have to put more sensors in the vehicle,

Morali bismo da postavimo više senzora u vozilo

and we'll pick some operating point up here

i izabraćemo jednu tačku korišćenja ovde

where it basically never gets into a crash.

gde, zapravo, nikada ne dolazi do sudara.

They'll happen, but very low frequency.

Dešavaće se, ali sa vrlo malom učestalošću.

Now you and I could look at this and we could argue

Mogli bismo ovo da gledamo i da polemišemo o tome

about whether it's incremental, and I could say something like "80-20 rule,"

da li je ovo postepeno i ja bih rekao da je to "pravilo 80-20"

and it's really hard to move up to that new curve.

i da je veoma teško da se podigne na tu novu krivu.

But let's look at it from a different direction for a moment.

Ali hajde na trenutak da to sagledamo iz drugačijeg ugla.

So let's look at how often the technology has to do the right thing.

Hajde da vidimo koliko često tehnologija postupi na ispravan način.

And so this green dot up here is a driver assistance system.

Ova zelena tačka ovde predstavlja sistem asistencije vozaču.

It turns out that human drivers

Ispostavilo se da ljudi vozači

make mistakes that lead to traffic accidents

prave greške koje dovode do saobraćajnih nesreća

about once every 100,000 miles in America.

jednom u 161 000 km, u Americi.

In contrast, a self-driving system is probably making decisions

Nasuprot tome, sistem asistencije vozaču verovatno donosi odluke

about 10 times per second,

oko 10 puta u sekundi,

so order of magnitude,

a to je red veličine

that's about 1,000 times per mile.

oko 1000 puta u toku 1,5 kilometra.

So if you compare the distance between these two,

Ako uporedite razdaljinu između ove dve veličine,

it's about 10 to the eighth, right?

to je oko 10 na osmu, zar ne?

Eight orders of magnitude.

Osam puta više.

That's like comparing how fast I run

To je kao kada bi poredili koliko brzo ja trčim

to the speed of light.

u odnosu na brzinu svetlosti.

It doesn't matter how hard I train, I'm never actually going to get there.

Nije bitno koliko naporno treniram kada zapravo, nikada neću moći to da stignem.

So there's a pretty big gap there.

Prema tome, to je vrlo velik procep.

And then finally, there's how the system can handle uncertainty.

I na kraju, naravno, imamo i to kako sistem podnosi nesigurnosti.

So this pedestrian here might be stepping into the road, might not be.

Ovaj pešak ovde možda iskorači na put, a možda i ne.

I can't tell, nor can any of our algorithms,

Ja ne mogu to da znam, niti bilo koji naš algoritam

but in the case of a driver assistance system,

ali u slučaju sistema asistencije vozaču

that means it can't take action, because again,

to znači da ne može nešto da preduzme opet iz razloga što

if it presses the brakes unexpectedly, that's completely unacceptable.

ako nenadano pritisne kočnice to je potpuno neprihvatljivo.

Whereas a self-driving system can look at that pedestrian and say,

Dok samoupravljajući sistem može da prepozna pešaka i kaže

I don't know what they're about to do,

"Ne znam šta će da uradi,

slow down, take a better look, and then react appropriately after that.

uspori, pogledaj bolje i nakon toga reaguj adekvatno."

So it can be much safer than a driver assistance system can ever be.

Tako da je mnogo bezbedniji nego što će sistem asistencije vozaču ikad biti.

So that's enough about the differences between the two.

Toliko o razlikama između ova dva sistema.

Let's spend some time talking about how the car sees the world.

Hajde da sada razgovaramo o tome kako automobil vidi svet.

So this is our vehicle.

Dakle, ovo je naše vozilo.

It starts by understanding where it is in the world,

Kreće tako što prepoznaje gde se nalazi, na svetu,

by taking a map and its sensor data and aligning the two,

tako što koristi podatke sa mape i senzora i upoređuje ih

and then we layer on top of that what it sees in the moment.

i na to dodaje šta vidi u konkretnom trenutku.

So here, all the purple boxes you can see are other vehicles on the road,

Tako da ove ljubičaste kutije koje vidite predstavljaju druga vozila na putu.

and the red thing on the side over there is a cyclist,

Ove crvene stvari, sa strane, su biciklisti,

and up in the distance, if you look really closely,

a tamo u daljini, ako bolje pogledate,

you can see some cones.

možete videti čunjeve.

Then we know where the car is in the moment,

Sad, pošto znamo gde se auto nalazi u datom trenutku

but we have to do better than that: we have to predict what's going to happen.

moramo da uradimo i bolje od toga: moramo da predvidimo šta će se desiti.

So here the pickup truck in top right is about to make a left lane change

Ovaj pikap kamion, gore desno, će da izvrši prestrojavanje u levu traku

because the road in front of it is closed,

zato što je napred put zatvoren

so it needs to get out of the way.

pa on mora da se skloni sa puta.

Knowing that one pickup truck is great,

Poznajemo taj jedan pikap i to je super

but we really need to know what everybody's thinking,

ali moramo da znamo i o čemu svi ostali misle

so it becomes quite a complicated problem.

pa to sad postaje vrlo komplikovan problem.

And then given that, we can figure out how the car should respond in the moment,

Ako sad to znamo, možemo da smislimo kako bi auto trebalo da reaguje u trenutku

so what trajectory it should follow, how quickly it should slow down or speed up.

koju putanju bi trebalo da sledi, koliko brzo bi trebalo da uspori ili ubrza.

And then that all turns into just following a path:

I sve to postaje samo praćenje putanje:

turning the steering wheel left or right, pressing the brake or gas.

skretanje levo ili desno, pritiskanje kočnice ili gasa.

It's really just two numbers at the end of the day.

Na kraju se sve svodi na dva broja, zapravo.

So how hard can it really be?

Koliko teško to može da bude?

Back when we started in 2009,

Kada smo počinjali u 2009,

this is what our system looked like.

ovako je izgledao naš sistem.

So you can see our car in the middle and the other boxes on the road,

Možete videti naš automobil u sredini i ostale kocke na putu

driving down the highway.

kako se voze po autoputu.

The car needs to understand where it is and roughly where the other vehicles are.

Auto mora da prepozna gde se nalazi i gde se otprilike nalaze i ostala vozila.

It's really a geometric understanding of the world.

To je zapravo geometrijsko razumevanje sveta.

Once we started driving on neighborhood and city streets,

Kada smo počeli da vozimo po komšiluku i gradskim ulicama,

the problem becomes a whole new level of difficulty.

pojavile su se poteškoće na potpuno novom nivou.

You see pedestrians crossing in front of us, cars crossing in front of us,

Vidimo pešake koji prelaze ispred nas, automobile koji prelaze ispred nas

going every which way,

koji idu u svim smerovima,

the traffic lights, crosswalks.

semafori, pešački prelazi.

It's an incredibly complicated problem by comparison.

To je neverovatno složen problem u odnosu na prethodni.

And then once you have that problem solved,

Jednom kada je taj problem rešen,

the vehicle has to be able to deal with construction.

vozilo mora da se nosi sa radovima na putu.

So here are the cones on the left forcing it to drive to the right,

Ovde su čunjevi sa leve strane koji ga primoravaju da ide na desnu

but not just construction in isolation, of course.

ali ne samo sa radovima, naravno.

It has to deal with other people moving through that construction zone as well.

Mora da se nosi sa ostalim ljudima koji se kreću oko tih radova, takođe.

And of course, if anyone's breaking the rules, the police are there

I, naravno, ako neko krši pravila, policija je tu

and the car has to understand that that flashing light on the top of the car

i auto mora da prepozna rotaciono svetlo na krovu auta

means that it's not just a car, it's actually a police officer.

koje znači da to nije običan auto, nego policijac.

Similarly, the orange box on the side here,

Slično, ova narandžasta kutija sa strane

it's a school bus,

je školski autobus

and we have to treat that differently as well.

i time moramo, takođe, drugačije da se pozabavimo.

When we're out on the road, other people have expectations:

Kada se nalazimo na putu, drugi ljudi imaju očekivanja.

So, when a cyclist puts up their arm,

Tako da kada biciklista ispruži ruku

it means they're expecting the car to yield to them and make room for them

to znači da očekuje od auta da mu da prednost i napravi mesta za njega

to make a lane change.

da bi se prestrojio.

And when a police officer stood in the road,

I kada se policijski auto nalazi na putu

our vehicle should understand that this means stop,

naše vozilo mora da prepozna da to znači da stanemo,

and when they signal to go, we should continue.

i kada nam signaliziraiju da krenemo da mi nastavimo dalje.

Now, the way we accomplish this is by sharing data between the vehicles.

Način na koji ovo postižemo jeste deljenjem podataka između vozila.

The first, most crude model of this

Prvi, grub model ovoga

is when one vehicle sees a construction zone,

je kada vozilo vidi radove,

having another know about it so it can be in the correct lane

i daje drugom to do znanja da bi moglo da bude u pravilnoj traci

to avoid some of the difficulty.

kako bi izbeglo poteškoće.

But we actually have a much deeper understanding of this.

Ovo zapravo mnogo dublje razumemo.

We could take all of the data that the cars have seen over time,

Mogli bismo da uzmemo sve podatke koje auto prima u toku vremena

the hundreds of thousands of pedestrians, cyclists,

stotine hiljada pešaka, motociklista,

and vehicles that have been out there

i vozila koje se tamo nalaze

and understand what they look like

i da prepoznamo kako izgledaju

and use that to infer what other vehicles should look like

i to upotrebimo da zaključimo kako bi druga vozila

and other pedestrians should look like.

i drugi pešaci mogli da izgledaju.

And then, even more importantly, we could take from that a model

I onda, još važnije, iz toga bismo mogli da izvučemo model

of how we expect them to move through the world.

kako očekujemo da se oni ponašaju u svetu.

So here the yellow box is a pedestrian crossing in front of us.

Ovde žuta kocka predstavlja pešaka koji prelazi ispred nas.

Here the blue box is a cyclist and we anticipate

Ovde je plava kocka biciklista i predviđamo

that they're going to nudge out and around the car to the right.

da će da se pomeri iza automobila desno.

Here there's a cyclist coming down the road

Ovde nam biciklista ide u susret

and we know they're going to continue to drive down the shape of the road.

i znamo da će nastaviti da vozi po obliku puta.

Here somebody makes a right turn,

Ovde neko skreće desno,

and in a moment here, somebody's going to make a U-turn in front of us,

i ubrzo vidimo ovde nekog ko će da napravi polukružno ispred nas

and we can anticipate that behavior and respond safely.

pa možemo da predvidimo to i da reagujemo bezbedno.

Now, that's all well and good for things that we've seen,

Sad, sve je to dobro za ove stvari koje smo videli.

but of course, you encounter lots of things that you haven't

Ali, naravno, dešavaće se gomila stvari

koje pre toga niste videli.

seen in the world before.

And so just a couple of months ago,

Pre nekoliko meseci,

our vehicles were driving through Mountain View,

naša vozila su se kretala kroz Mauntin Vju

and this is what we encountered.

i evo sa čim smo se susreli.

This is a woman in an electric wheelchair

Ovo je žena u električnim kolicima

chasing a duck in circles on the road. (Laughter)

koja juri patku u krug, po putu. (Smeh)

Now it turns out, there is nowhere in the DMV handbook

Međutim, nigde u uputstvu o ponašanju u saobraćaju

that tells you how to deal with that,

ne piše kako to da rešite.

but our vehicles were able to encounter that,

Ali naša vozila su mogla to da prepoznaju,

slow down, and drive safely.

uspore i nastave bezbedno.

Now, we don't have to deal with just ducks.

Nećemo se nositi samo sa patkama.

Watch this bird fly across in front of us. The car reacts to that.

Vidite kako ova ptica izleće ispred nas.

Auto reaguje na to.

Here we're dealing with a cyclist

Ovde se suočavamo sa biciklistom

that you would never expect to see anywhere other than Mountain View.

kojeg ne biste očekivali da vidite nigde drugde osim u Mauntin Vjuu.

And of course, we have to deal with drivers,

I, naravno, moramo da se suočavamo

even the very small ones.

čak i sa veoma malim vozačima.

Watch to the right as someone jumps out of this truck at us.

Pogledajte desno ovde kako neko iskače iz ovog kamiona ispred nas.

And now, watch the left as the car with the green box decides

I ovde sa leve strane, auto sa zelenom kutijom koji je odlučio

he needs to make a right turn at the last possible moment.

da mora da skrene desno u poslednjem mogućem trenutku.

Here, as we make a lane change, the car to our left decides

Ovde, dok se prestrojavamo, auto sa naše leve strane

it wants to as well.

želi isto to.

And here, we watch a car blow through a red light

A ovde vidimo auto koji jurca kroz crveno svetlo

and yield to it.

i propuštamo ga.

And similarly, here, a cyclist blowing through that light as well.

Slično tome, ovde biciklista isto jurca kroz crveno svetlo.

And of course, the vehicle responds safely.

I, naravno, vozilo reaguje bezbedno.

And of course, we have people who do I don't know what

Takođe imamo ljude koji rade ne-znam-ni-ja-šta na putu

sometimes on the road, like this guy pulling out between two self-driving cars.

kao ovaj koji se zaustavlja između dva samoupravljajuće automobila.

You have to ask, "What are you thinking?"

Zapitate se: "O čemu ti razmišljaš?"

(Laughter)

(Smeh)

Now, I just fire-hosed you with a lot of stuff there,

Obasuo sam vas sa dosta stvari

so I'm going to break one of these down pretty quickly.

i sada ću samo na brzinu analizirati jednu od njih.

So what we're looking at is the scene with the cyclist again,

Sada gledamo situaciju, opet sa biciklistom,

and you might notice in the bottom, we can't actually see the cyclist yet,

i na dnu možete da primetite da još uvek ne možemo da ga vidimo

but the car can: it's that little blue box up there,

ali auto može: to je ona mala plava kocka tamo,

and that comes from the laser data.

i to je podatak koji se očitao laserski.

And that's not actually really easy to understand,

To nije baš tako lako da se razume

so what I'm going to do is I'm going to turn that laser data and look at it,

pa ću ih sada uključiti da vidimo te laserske podatke

pa ako ste veoma dobri u gledanju laserskih podataka,

and if you're really good at looking at laser data, you can see

možete uočiti ove tačke na bankini

a few dots on the curve there,

right there, and that blue box is that cyclist.

ovde, a ona plava kocka je taj biciklista.

Pošto je nama sada crveno svetlo,

Now as our light is red,

the cyclist's light has turned yellow already,

biciklisti se već upalilo žuto,

and if you squint, you can see that in the imagery.

i ako škiljite, možete da vidite to ovde na slici.

But the cyclist, we see, is going to proceed through the intersection.

Ali biciklista, kako vidimo, će nastaviti pravo na raskrsnici.

Our light has now turned green, his is solidly red,

Nama je sada zeleno svetlo, njegovo je crveno,

and we now anticipate that this bike is going to come all the way across.

i mi sada predviđamo da će taj bajs da pređe preko.

Unfortunately the other drivers next to us were not paying as much attention.

Nažalost, ostali vozači pored nas nisu obratili dovoljno pažnje.

They started to pull forward, and fortunately for everyone,

Počeli su da se pomeraju unapred i na svu sreću,

this cyclists reacts, avoids,

biciklista je reagovao, izbegao,

and makes it through the intersection.

i prošao kroz raskrsnicu.

And off we go.

I krećemo.

Now, as you can see, we've made some pretty exciting progress,

Kao što vidite, napravili smo prilično uzbudljiv napredak

and at this point we're pretty convinced

i u ovom trenutku smo prilično ubeđeni

this technology is going to come to market.

da će se ova tehnologija naći na tržištu.

We do three million miles of testing in our simulators every single day,

U našim simulatorima testiramo skoro 5 miliona kilometara svakog dana

so you can imagine the experience that our vehicles have.

pa možete zamisliti kakvo iskustvo naša vozila imaju.

We are looking forward to having this technology on the road,

Radujemo se što će ova tehnologija biti na putu,

and we think the right path is to go through the self-driving

i mislimo da su samoupravljajući automobili pravi put

rather than driver assistance approach

u odnosu na sistem asistencije vozaču

because the urgency is so large.

zato što je pravovremenost veoma značajna.

In the time I have given this talk today,

U toku ovog mog današnjeg govora

34 people have died on America's roads.

34 ljudi je poginulo na američkim putevima.

How soon can we bring it out?

Koliko brzo možemo izneti ovu tehnologiju?

Well, it's hard to say because it's a really complicated problem,

Teško je to reći jer je to zaista komplikovan problem.

but these are my two boys.

Ovo su moja dva dečaka.

My oldest son is 11, and that means in four and a half years,

Stariji ima 11 godina a to znači da će za četiri ipo godine

he's going to be able to get his driver's license.

moći da dobije vozačku dozvolu.

My team and I are committed to making sure that doesn't happen.

Moj tim i ja smo posvećeni tome da se to ne desi.

Thank you.

Hvala vam.

(Laughter) (Applause)

(Smeh) (Aplauz)

Chris Anderson: Chris, I've got a question for you.

Kris Anderson: Kris, imam pitanje za tebe.

Chris Urmson: Sure.

Kris Urmson: Svakako.

CA: So certainly, the mind of your cars is pretty mind-boggling.

KA: Zasigurno, svest vaših automobila je prilično zapanjujuća.

On this debate between driver-assisted and fully driverless --

Na debatu između asistencije vozaču i potpuno bezvozača -

I mean, there's a real debate going on out there right now.

Mislim, dešava se prava debata o tome sada.

So some of the companies, for example, Tesla,

Pa neke kompanije, kao što je, na primer, Tesla

are going the driver-assisted route.

idu putem asistencije vozaču.

What you're saying is that that's kind of going to be a dead end

Ti govoriš da će to biti ćorsokak

because you can't just keep improving that route and get to fully driverless

jer ne može samo unapređivanjem te putanje

da se dođe do potpunog samoupravljanja

at some point, and then a driver is going to say, "This feels safe,"

u jednom trenutku i da onda vozač kaže: "Osećam da je ovo sigurno"

and climb into the back, and something ugly will happen.

i onda se prebaci nazad i nešto ružno se desi.

CU: Right. No, that's exactly right, and it's not to say

KU: Tako je. To je potpuno tačno i ne možemo da kažemo

that the driver assistance systems aren't going to be incredibly valuable.

da sistemi asistencije vozaču neće biti neverovatno značajni.

They can save a lot of lives in the interim,

Mogu da spasu mnogo života u međuvremenu,

but to see the transformative opportunity to help someone like Steve get around,

ali da imaju priliku da se transformišu i pomognu ljudima kao što je Stiv

to really get to the end case in safety,

da zaista dođu do kraja bezbedno,

to have the opportunity to change our cities

da imaju mogućnost da promene naše gradove

and move parking out and get rid of these urban craters we call parking lots,

i da se otarase urbanih kratera koje zovemo parking mestima,

it's the only way to go.

ovo je jedini put.

CA: We will be tracking your progress with huge interest.

KA: Pratićemo vaš napredak sa velikim interesovanjem.

Thanks so much, Chris. CU: Thank you. (Applause)

Hvala puno, Kris. KU: Hvala vama. (Aplauz)