Chris Urmson: How a driverless car sees the road
Chris Urmson: Jak samosterujący samochód widzi drogę
Double-click the English transcript below to play the video.
invented the automobile.
wymyślił automobil.
for the first public test drive,
pierwszą publiczną jazdę próbną
crashed into a wall.
reliable part of the car, the driver.
częścią samochodu, kierowcą.
we've added air bags,
i poduszki powietrzne.
started trying to make the car smarter
uczynić samochody mądrzejsze,
a little bit about the difference
with driver assistance systems
systemami wspomagania kierowcy
self-driving cars
samosterujących samochodów
a little bit about our car
o naszym samochodzie
and how it reacts and what it does,
reaguje i działa,
a little bit about the problem.
on the world's roads every year.
ginie 1,2 miliona ludzi.
are killed each year.
33 tysięcy osób rocznie.
falling out of the sky every working day.
spadał Boeing 737.
other than drive.
innego niż jazdę.
increased by 38 percent.
samochody wzrósł o 38%.
than it was not very long ago.
dużo gorszy niż jeszcze parę lat temu.
in America, which is about 50 minutes,
dojazdu do pracy w Ameryce, 50 minut,
workers we have,
about six billion minutes
każdego dnia.
so let's put it in perspective.
life expectancy of a person,
przewidywaną długość życia
who don't have the privilege
którzy nie mają przywileju
drive to work in the morning,
30-minutowej jazdy do pracy
of piecing together bits of public transit
z komunikacją publiczną
o podwózkę.
that you and I have to get around.
co ja i ty.
these driver assistance systems
and incrementally improve them,
wspomagania kierowcy,
into self-driving cars.
w samosterujące samochody.
that's like me saying
one day I'll be able to fly.
something a little different.
about three different ways
than driver assistance systems.
a wspomagającymi.
with some of our own experience.
of a self-driving car
samosterującym samochodem
they were 100 Googlers,
pracownikom Google,
them to use it in their daily lives.
this one had a big asterisk with it:
but it could still fail.
mimo że był po testach.
we let them use it,
was something awesome,
to bring a product into the world.
who came in and told us on the first day,
który pierwszego dnia powiedział:
What are we thinking?"
Co wy myślicie?".
"Not only should I have it,
"Nie tylko ja powinienem go mieć,
because people are terrible drivers."
bo ludzie to beznadziejni kierowcy".
the people inside the car were doing,
co ludzie robili w samochodzie
and realizes the battery is low,
i widzi, że bateria jest słaba,
and digs around in his backpack,
i grzebie w plecaku,
the charging cable for his phone,
puts it on the phone.
podłącza ją do laptopa i telefonu.
65 miles per hour down the freeway.
100 na godzinę autostradą.
it's kind of obvious, right?
uznaliśmy to za oczywiste,
the driver is going to get.
incrementally smarter,
coraz mądrzejszych pojazdów
the wins we really need.
czego chcemy.
a little technical for a moment here.
and along the bottom
apply the brakes when it shouldn't.
and the car starts stopping randomly,
nagle zatrzymywać,
the car is going to apply the brakes
samochód używa hamulców,
to help you avoid an accident.
the bottom left corner here,
it doesn't do anything goofy,
nie zrobi nic głupiego,
out of an accident.
a driver assistance system into a car,
system wspomagający
unikającym kolizji,
of technology on there,
to have some operating properties,
Będzie to funkcjonalne,
all of the accidents,
wszystkim wypadkom,
along the curve here,
punkt na krzywej,
that the human driver misses,
jakie miałby kierowca.
by a factor of two.
wypadków.
dying every year in America.
ofiar drogowych co roku.
that looks like this.
musi wyglądać tak.
more sensors in the vehicle,
więcej czujników,
operating point up here
gets into a crash.
and we could argue
czy to przyrost,
I could say something like "80-20 rule,"
to that new curve.
dojść do nowej krzywej.
from a different direction for a moment.
z innej perspektywy.
the technology has to do the right thing.
musi robić dobre rzeczy.
is a driver assistance system.
to system wspomagający.
to traffic accidents
is probably making decisions
prawdopodobnie decyduje
between these two,
szybkość mojego biegu
I'm never actually going to get there.
i tak go nie dogonię.
the system can handle uncertainty.
radzi sobie z niepewnością.
stepping into the road, might not be.
albo nie.
nor can any of our algorithms,
ani żaden algorytm,
a driver assistance system,
because again,
that's completely unacceptable.
jest niedopuszczalne.
can look at that pedestrian and say,
widzi przechodnia i myśli:
and then react appropriately after that.
i odpowiednio zareaguję."
a driver assistance system can ever be.
niż jakikolwiek system wspomagający.
the differences between the two.
how the car sees the world.
where it is in the world,
and aligning the two,
i danych z sensorów,
what it sees in the moment.
aktualny widok.
are other vehicles on the road,
to inne pojazdy,
over there is a cyclist,
if you look really closely,
is in the moment,
jest w tej chwili,
we have to predict what's going to happen.
przewidywać, co się stanie.
is about to make a left lane change
zmienia pas na lewy,
what everybody's thinking,
zamiary wszystkich,
how the car should respond in the moment,
w danej chwili,
quickly it should slow down or speed up.
jak szybko jechać.
just following a path:
do jazdy ścieżką:
pressing the brake or gas.
naciśnięcie gazu lub hamulca.
at the end of the day.
and the other boxes on the road,
i inne obiekty,
and roughly where the other vehicles are.
i gdzie są inne pojazdy.
understanding of the world.
on neighborhood and city streets,
do osiedli i miast,
new level of difficulty.
of us, cars crossing in front of us,
oraz inne auta,
i przejścia dla pieszych.
problem by comparison.
problem.
that problem solved,
to deal with construction.
przejechać przez teren budowy.
forcing it to drive to the right,
zjechać na prawo,
in isolation, of course.
through that construction zone as well.
z przechodniami.
breaking the rules, the police are there
pojawia się policja,
that flashing light on the top of the car
że błyskające światło
it's actually a police officer.
ale policja.
on the side here,
differently as well.
innego traktowania.
other people have expectations:
to yield to them and make room for them
iż samochód ustąpi
stood in the road,
that this means stop,
że oznacza to stop,
we should continue.
może jechać.
is by sharing data between the vehicles.
wymianie danych między pojazdami.
sees a construction zone,
so it can be in the correct lane
deeper understanding of this.
that the cars have seen over time,
z samochodu,
of pedestrians, cyclists,
what other vehicles should look like
jak mogą wyglądać inne pojazdy
we could take from that a model
możemy stworzyć model oczekiwań
to move through the world.
uczestników ruchu.
crossing in front of us.
przechodzący przed nami.
and we anticipate
and around the car to the right.
że samochód minie go po prawej.
coming down the road
to drive down the shape of the road.
jechał wzdłuż drogi.
going to make a U-turn in front of us,
and respond safely.
i bezpiecznie zareagować.
for things that we've seen,
lots of things that you haven't
through Mountain View,
jechały przez Mountain View
(Laughter)
(Śmiech)
in the DMV handbook
to encounter that,
with just ducks.
The car reacts to that.
i reakcję samochodu.
anywhere other than Mountain View.
oczekiwanym miejscu.
to deal with drivers,
też z kierowcami,
jumps out of this truck at us.
na nas z ciężarówki.
with the green box decides
zielony samochód uznaje,
at the last possible moment.
w ostatniej chwili.
the car to our left decides
samochód po lewej uznał,
blow through a red light
na czerwonym świetle,
blowing through that light as well.
też przejeżdża na czerwonym.
the vehicle responds safely.
reaguje bezpiecznie.
who do I don't know what
pulling out between two self-driving cars.
pomiędzy dwa nasze auta.
with a lot of stuff there,
down pretty quickly.
with the cyclist again,
we can't actually see the cyclist yet,
blue box up there,
ten mały niebieski obiekt
really easy to understand,
to turn that laser data and look at it,
at laser data, you can see
zauważycie
is that cyclist.
has turned yellow already,
in the imagery.
to proceed through the intersection.
przez skrzyżowanie.
his is solidly red,
on czerwone,
is going to come all the way across.
przejedzie w poprzek drogi.
were not paying as much attention.
nie zwracają na to uwagi
and fortunately for everyone,
some pretty exciting progress,
duży postęp
to come to market.
in our simulators every single day,
testów dziennie,
that our vehicles have.
this technology on the road,
tej technologii na drodze.
is to go through the self-driving
jest przejście na automatykę
it's a really complicated problem,
bo to skomplikowany problem.
in four and a half years,
więc za 4,5 roku
to get his driver's license.
to making sure that doesn't happen.
by do tego nie dopuścić.
(Brawa)
I've got a question for you.
mam do ciebie pytanie.
is pretty mind-boggling.
jest niepojęty.
driver-assisted and fully driverless --
going on out there right now.
for example, Tesla,
na przykład Tesla,
that's kind of going to be a dead end
that route and get to fully driverless
nie dojdą do pełnej automatyki?
is going to say, "This feels safe,"
and something ugly will happen.
i stanie się coś złego.
and it's not to say
aren't going to be incredibly valuable.
jest bezwartościowe.
in the interim,
to help someone like Steve get around,
tak by ludzie jak Steve
to change our cities
these urban craters we call parking lots,
with huge interest.
będziemy śledzić wasze postępy.
CU: Thank you. (Applause)
CU: Dziękuję. (Brawa)
ABOUT THE SPEAKER
Chris Urmson - RoboticistChris Umson is the Director of Self-Driving Cars at Google[x].
Why you should listen
Since 2009, Chris Urmson has headed up Google’s self-driving car program. So far, the team’s vehicles have driven over three quarters of a million miles. While early models included a driverless Prius that TEDsters got to test- ... um, -not-drive in 2011, more and more the team is building vehicles from the ground up, custom-made to go driverless.
Prior to joining Google, Umson was on the faculty of the Robotics Institute at Carnegie Mellon University, where his research focused on motion planning and perception for robotic vehicles. During his time at Carnegie Mellon, he served as Director of Technology for the team that won the 2007 DARPA Urban Challenge.
Chris Urmson | Speaker | TED.com