Chris Urmson: How a driverless car sees the road
크리스 엄슨(Chris Urmson): 무인자동차가 도로를 보는 방법
Double-click the English transcript below to play the video.
invented the automobile.
자동차를 발명했습니다.
for the first public test drive,
앞에서 시험 운행을 했습니다.
crashed into a wall.
reliable part of the car, the driver.
운전자에 대해서 연구를 했습니다.
we've added air bags,
started trying to make the car smarter
만들기 위한 시도를 시작했습니다.
a little bit about the difference
with driver assistance systems
이야기 하고자 합니다.
self-driving cars
a little bit about our car
잠시 이야기 할 예정입니다.
and how it reacts and what it does,
반응하며 실제로 무엇인지 말이죠.
a little bit about the problem.
관해 이야기 할까 합니다.
on the world's roads every year.
전세계의 도로에서 죽습니다.
are killed each year.
사람들이 매해 죽고 있습니다.
falling out of the sky every working day.
한 대가 추락하는 것과 같습니다.
비오는 날씨에는
other than drive.
다른 것이 하고 싶을 겁니다.
increased by 38 percent.
than it was not very long ago.
그리 오래되지는 않았습니다.
인적비용을 초래합니다.
in America, which is about 50 minutes,
평균이동시간인 50분을
workers we have,
사람에 곱하게 되면
about six billion minutes
결론이 나옵니다.
so let's put it in perspective.
조금 다른 관점에서 보면,
life expectancy of a person,
who don't have the privilege
drive to work in the morning,
of piecing together bits of public transit
대중교통을 이용하거나
픽업을 부탁해야 합니다.
that you and I have to get around.
가지고 있는 자유가 없습니다.
these driver assistance systems
and incrementally improve them,
into self-driving cars.
개발하는 것 입니다.
that's like me saying
one day I'll be able to fly.
언젠가 하늘을 날 수 있을꺼야"
something a little different.
할 필요가 있습니다.
about three different ways
운전보조 시스템이
than driver assistance systems.
이야기해보고자 합니다.
with some of our own experience.
이야기하며 시작할까 합니다.
of a self-driving car
첫 시범 운행을 했습니다.
they were 100 Googlers,
them to use it in their daily lives.
일상 생활에 사용하도록 했습니다.
this one had a big asterisk with it:
크게 주목해야 할 점이 있었습니다.
차량이었으니까요.
but it could still fail.
계속 실패했습니다.
we let them use it,
was something awesome,
to bring a product into the world.
소개하려고 시도할 때처럼
who came in and told us on the first day,
첫 날 이런 말을 했습니다.
What are we thinking?"
"Not only should I have it,
"나만 가질 것이 아니라,
because people are terrible drivers."
라고 말했습니다.
the people inside the car were doing,
무슨 일을 하는지 살펴보았습니다.
한 신사분의 이야기입니다.
and realizes the battery is low,
얼마 남지 않았다는 것을 알았습니다.
and digs around in his backpack,
그의 가방안을 뒤집니다.
the charging cable for his phone,
puts it on the phone.
65 miles per hour down the freeway.
달리는 동안에 말이죠.
it's kind of obvious, right?
the driver is going to get.
incrementally smarter,
똑똑하게 만드는 것으로는
the wins we really need.
것을 볼 수 없습니다.
a little technical for a moment here.
약간 설명하겠습니다.
and along the bottom
apply the brakes when it shouldn't.
얼마나 밟았나 보여줍니다.
and the car starts stopping randomly,
자동차는 어느때나 멈출테니까요.
the car is going to apply the brakes
얼마나 자주 브레이크를
to help you avoid an accident.
the bottom left corner here,
it doesn't do anything goofy,
out of an accident.
시스템도 없습니다.
a driver assistance system into a car,
자동차에 추가시키기 위해
of technology on there,
것들을 넣어볼까 합니다.
to have some operating properties,
적용할때 나타날 그래프입니다.
예방할 수는 없습니다.
all of the accidents,
along the curve here,
중심으로 다루어볼까합니다.
that the human driver misses,
절반의 사고를 줄여줍니다.
by a factor of two.
절반으로 줄일 수 있습니다.
dying every year in America.
미국에서 매해 사망할 것입니다.
that looks like this.
모양으로 보여야합니다.
more sensors in the vehicle,
삽입할 것입니다.
operating point up here
개선시킬 것입니다.
gets into a crash.
매우 낮은 횟수일 것입니다.
and we could argue
가능한지 이야기 해봅시다.
I could say something like "80-20 rule,"
이라고 말합니다.
to that new curve.
것은 정말 힘듭니다.
from a different direction for a moment.
the technology has to do the right thing.
is a driver assistance system.
운전보조 시스템입니다.
to traffic accidents
실수를 하는 것을 보여줍니다.
is probably making decisions
between these two,
이 두가지를 비교한다면,
한 것입니다. 그렇죠?
I'm never actually going to get there.
저는 절대 따라잡을 수 없을 것입니다.
the system can handle uncertainty.
다루는 지 알 수 있습니다.
stepping into the road, might not be.
있고 안 갈 수도 있습니다.
nor can any of our algorithms,
알고리즘도 마찬가지 입니다.
a driver assistance system,
because again,
that's completely unacceptable.
용납할 수 없는 일이기 때문입니다.
can look at that pedestrian and say,
보행자를 볼 수있고
and then react appropriately after that.
그 후에 적절한 행동을 할 것입니다.
a driver assistance system can ever be.
더욱 안전할 수 밖에 없습니다.
the differences between the two.
차이에 대해서 설명했습니다.
how the car sees the world.
대해 이야기 나눠보도록 합시다.
where it is in the world,
감지하면서 시작합니다.
and aligning the two,
what it sees in the moment.
주변상황을 삽입합니다.
are other vehicles on the road,
다른 차들을 의미하고
over there is a cyclist,
if you look really closely,
is in the moment,
we have to predict what's going to happen.
키워야 합니다.
is about to make a left lane change
앞쪽 도로가 막혀서
what everybody's thinking,
생각을 읽어야 합니다.
how the car should respond in the moment,
순간적으로 판단해야 합니다.
quickly it should slow down or speed up.
얼마나 속도를 조절할 것인가
just following a path:
결론으로 귀결됩니다.
pressing the brake or gas.
밟느냐 안밟으냐 입니다.
at the end of the day.
and the other boxes on the road,
and roughly where the other vehicles are.
다른 차량들의 위치도 알아야 합니다.
understanding of the world.
on neighborhood and city streets,
new level of difficulty.
of us, cars crossing in front of us,
신호등, 횡단보도가 나타났습니다.
problem by comparison.
복잡한 문제였죠.
that problem solved,
to deal with construction.
대처해야 했죠.
forcing it to drive to the right,
차를 유도합니다.
in isolation, of course.
문제들도 존재합니다.
through that construction zone as well.
움직이는 것도 고려해야하죠.
breaking the rules, the police are there
생기면 경찰이 출동할 수 있고
that flashing light on the top of the car
불빛이 반짝이는 차는
it's actually a police officer.
on the side here,
differently as well.
other people have expectations:
to yield to them and make room for them
stood in the road,
뜻임을 알아채야 합니다.
that this means stop,
we should continue.
물론 알아야 겠죠.
is by sharing data between the vehicles.
공유함으로써 해결책을 찾았습니다.
sees a construction zone,
so it can be in the correct lane
deeper understanding of this.
대해 잘 이해시켰습니다.
that the cars have seen over time,
of pedestrians, cyclists,
what other vehicles should look like
we could take from that a model
to move through the world.
crossing in front of us.
앞을 지나는 보행자들입니다.
and we anticipate
and around the car to the right.
것이라는 것을 예상합니다.
coming down the road
to drive down the shape of the road.
것을 알고 있습니다.
going to make a U-turn in front of us,
앞에서 유턴하고 있습니다.
안전하게 반응합니다.
and respond safely.
for things that we've seen,
lots of things that you haven't
through Mountain View,
운행했습니다.
(Laughter)
빙빙돌고 있었습니다.
in the DMV handbook
to encounter that,
with just ducks.
The car reacts to that.
무인자동차는 반응합니다.
anywhere other than Mountain View.
예상하기 어렵겠죠.
to deal with drivers,
jumps out of this truck at us.
잠깐 내리는걸 보세요.
with the green box decides
우회전 하는 차량을 보세요.
at the last possible moment.
the car to our left decides
blow through a red light
blowing through that light as well.
the vehicle responds safely.
who do I don't know what
사람들이 있습니다.
pulling out between two self-driving cars.
본인 차를 집어 넣었습니다.
"무슨 생각 인가요?"
with a lot of stuff there,
down pretty quickly.
with the cyclist again,
시각에서 자전거가 보이지 않습니다.
we can't actually see the cyclist yet,
blue box up there,
really easy to understand,
to turn that laser data and look at it,
at laser data, you can see
곡선상의 몇개의 점들이 보입니다.
is that cyclist.
has turned yellow already,
in the imagery.
to proceed through the intersection.
교차로에 진입했습니다.
자전거등은 빨간등이 됐습니다.
his is solidly red,
is going to come all the way across.
were not paying as much attention.
그 사실을 몰랐습니다.
and fortunately for everyone,
흥미로운 진행상황을 보였습니다.
some pretty exciting progress,
to come to market.
in our simulators every single day,
주행을 매일 하고 있습니다.
that our vehicles have.
상상할 수 있겠죠.
적용되기를 고대하고 있습니다.
this technology on the road,
is to go through the self-driving
도로에서 사망합니다.
it's a really complicated problem,
문제라서 알 수 없습니다.
in four and a half years,
to get his driver's license.
딸 수 있습니다.
to making sure that doesn't happen.
않기 위해서 노력합니다.
I've got a question for you.
크리스, 질문이 있습니다.
is pretty mind-boggling.
생각은 무척 놀라운데요.
driver-assisted and fully driverless --
대한 토론들이 많습니다.
going on out there right now.
많이 진행중인데요.
for example, Tesla,
that's kind of going to be a dead end
막다른 길에 다다를 것이며
that route and get to fully driverless
무인자동차를 만들 수 없으며
is going to say, "This feels safe,"
and something ugly will happen.
and it's not to say
aren't going to be incredibly valuable.
없는 일은 아닙니다.
in the interim,
to help someone like Steve get around,
to change our cities
these urban craters we call parking lots,
도시 분화구를 사라지게할 방법은
with huge interest.
CU: Thank you. (Applause)
고맙습니다. 크리스씨
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