Chris Gerdes: The future race car -- 150mph, and no driver

Translator: Morton Bast Reviewer: Thu-Huong Ha

譯者: Jonas Lau 審譯者: Tien-Hsin Yang

So, how many of you have ever

在你們當中有多少人曾經

gotten behind the wheel of a car

在你根本不應該駕駛的狀況下

when you really shouldn't have been driving?

坐在駕駛者的位上呢?

Maybe you're out on the road for a long day,

也許你曾經一天開車很久了

and you just wanted to get home.

當下你只想回家

You were tired, but you felt you could drive a few more miles.

你非常累，但是你覺得你還可以再開幾哩路

Maybe you thought,

當下你也許在想,

I've had less to drink than everybody else,

我喝的酒比其他人都還少

I should be the one to go home.

應是我開車回家吧

Or maybe your mind was just entirely elsewhere.

又或者你的注意力完全不在開車這件事上了

Does this sound familiar to you?

這些想法聽起來很熟悉吧?

Now, in those situations, wouldn't it be great

現在，在剛剛所說的那些狀況下， 你應該會感到以下的新發明很吸引吧...

if there was a button on your dashboard

如果有一個按鈕在你的儀表板上，按一按

that you could push, and the car would get you home safely?

車子會自動地把你安全地送回家

Now, that's been the promise of the self-driving car,

這是自動導航汽車的承諾

the autonomous vehicle, and it's been the dream

一輛自主的汽車，這是從一九三九年

since at least 1939, when General Motors showcased

由通用汽車公司

this idea at their Futurama booth at the World's Fair.

在世界博覽會“飛出個未來"攤位中展示他們的想法開始後，很多人對汽車的寄望

Now, it's been one of those dreams

現在, 這些夢想

that's always seemed about 20 years in the future.

好像永遠要在二十年後才有希望實現

Now, two weeks ago, that dream took a step forward,

現在, 就在兩星期前, 這個夢想向前邁進了一步

when the state of Nevada granted Google's self-driving car

當美國內華達州對 Google (谷歌) 的自動導航汽車發出

the very first license for an autonomous vehicle,

首張針對自動汽車的行車證

clearly establishing that it's legal for them

明確確立法洌上容許這輛車

to test it on the roads in Nevada.

可以在內華達州的道路上進行測試

Now, California's considering similar legislation,

最近, 加州正考慮仿效類似的立法工作

and this would make sure that the autonomous car

這樣一來, 他們就可以確保那輛自動汽車

is not one of those things that has to stay in Vegas.

不會只在內華達州出現

(Laughter)

(笑聲)

Now, in my lab at Stanford, we've been working on

在我的史丹福大學實驗室裏, 我們也在著手研究

autonomous cars too, but with a slightly different spin

自動汽車, 但是從一個稍為不同的角度

on things. You see, we've been developing robotic race cars,

去看這件事。你看, 我們正在研發機動賽車

cars that can actually push themselves to the very limits

它們可以把自己的物理表現

of physical performance.

推至極限

Now, why would we want to do such a thing?

為甚麼我們會想做這件事呢?

Well, there's two really good reasons for this.

嗯, 這裏有兩個非常好的原因

First, we believe that before people turn over control

第一, 我們相信, 在人類把控制權

to an autonomous car, that autonomous car should be

交給自動汽車之前, 那自動汽車的表現至少要

at least as good as the very best human drivers.

能夠與最好的人類駕駛者匹敵

Now, if you're like me, and the other 70 percent of the population

如果你跟我一樣, 跟百分之七十的人口一樣

who know that we are above-average drivers,

都知道我們比一般駕駛者優勝

you understand that's a very high bar.

你會明白這對於自動汽車來說是很高的門檻

There's another reason as well.

還有另一個原因

Just like race car drivers can use all of the friction

就像賽車手能夠充份運用

between the tire and the road,

輪胎與路面之間的摩擦力

all of the car's capabilities to go as fast as possible,

加上充份運用車子的性能去達致最高速

we want to use all of those capabilities to avoid

我們想運用那些性質

any accident we can.

去減少任何意外的發生

Now, you may push the car to the limits

有時你使車子超過所能承受的極限

not because you're driving too fast,

未必因為你把車子開得太快

but because you've hit an icy patch of road,

而是因為你碰到地上的冰塊

conditions have changed.

以致於狀況有所改變。

In those situations, we want a car

在那些情況, 我們想要有

that is capable enough to avoid any accident

只要是物理上能被避免的

that can physically be avoided.

就能購避免任何意外發生的一輛車。

I must confess, there's kind of a third motivation as well.

我必須承認, 我們有第三個動機

You see, I have a passion for racing.

你看, 我有一股對賽車的熱誠

In the past, I've been a race car owner,

以前, 我曾經是一輛賽車車主

a crew chief and a driving coach,

一個車隊的領隊及駕駛教練

although maybe not at the level that you're currently expecting.

雖然不是你們想像中的那個層次

One of the things that we've developed in the lab --

我們在實驗室裏研發的其中一個項目--

we've developed several vehicles --

我們研發過幾款汽車--

is what we believe is the world's first

是我們認為是世界上第一輛

autonomously drifting car.

會飄移的自動汽車

It's another one of those categories

這是一個比較少

where maybe there's not a lot of competition.

競爭對手的研究範疇

(Laughter)

(笑聲)

But this is P1. It's an entirely student-built electric vehicle,

這是 P1。一個完全由學生建立的電動汽車

which through using its rear-wheel drive

它用它的尾輪

and front-wheel steer-by-wire

以及電子操控的前輪

can drift around corners.

可以在車道的彎位飄移

It can get sideways like a rally car driver,

它可以像拉力賽中的賽車手高速轉向

always able to take the tightest curve,

永遠能夠在最窄的角度行走

even on slippery, changing surfaces,

即使賽道濕滑、路面情況轉變

never spinning out.

它都不會滑出賽道

We've also worked with Volkswagen Oracle,

我們也跟福特汽車合作

on Shelley, an autonomous race car that has raced

用它的Shelley 型號汽車, 一款自動賽車

at 150 miles an hour through the Bonneville Salt Flats,

以時速150 英里穿越美國猶他州邦納維爾大鹽湖附近的平原

gone around Thunderhill Raceway Park in the sun,

在太陽底下橫過 Thunderhill 賽車場

the wind and the rain,

也在大風和雨中行走過該處

and navigated the 153 turns and 12.4 miles

以導航系統引領汽車在科羅拉多州的 Pikes Peak Hill Climb 賽道上

of the Pikes Peak Hill Climb route

轉153 個圈及走12.4 英里

in Colorado with nobody at the wheel.

汽車是無人駕駛的

(Laughter)

(笑聲)

(Applause)

(掌聲)

I guess it goes without saying that we've had a lot of fun

我毋須多說，這些經歷為我們

doing this.

帶來無窮的樂趣

But in fact, there's something else that we've developed

但實際上，我們在研發了

in the process of developing these autonomous cars.

這些自動汽車的過程中，體驗到一些事情

We have developed a tremendous appreciation

我們培養出一種對人類賽車手

for the capabilities of human race car drivers.

的能力的高度欣賞

As we've looked at the question of how well do these cars perform,

當我們看到這些自動汽車的表現時

we wanted to compare them to our human counterparts.

我們想將它們與人類賽車手作比較

And we discovered their human counterparts are amazing.

我們發現人類賽車手的表現是多麼令人讚歎的

Now, we can take a map of a race track,

如果我們拿到賽道的地圖

we can take a mathematical model of a car,

我們可以為自動汽車製作一個數學模型

and with some iteration, we can actually find

經過幾個回合的修定，我們的確可以找到

the fastest way around that track.

穿越賽道的最快路徑

We line that up with data that we record

我們將這些記錄得來的數據

from a professional driver,

與專業的賽車手比較

and the resemblance is absolutely remarkable.

結果兩者的吻合實在值得靈人注意

Yes, there are subtle differences here,

是的，兩者有少許的差異

but the human race car driver is able to go out

但人類賽車手能夠出賽

and drive an amazingly fast line,

然後奇妙的高速行駛

without the benefit of an algorithm that compares

卻毋須任何演算法去計算

the trade-off between going as fast as possible

車子以最高速轉向

in this corner, and shaving a little bit of time

以及不在直路上行走所損失的時間

off of the straight over here.

兩者之間的取捨

Not only that, they're able to do it lap

不單如此，他們能夠在一在一圈

after lap after lap.

又一圈... 又一圈後做出同樣的決定

They're able to go out and consistently do this,

他們在賽道上的表現非常穩定

pushing the car to the limits every single time.

每一次都把車子的性能發揮到極限

It's extraordinary to watch.

這的確令人歎為觀止

You put them in a new car,

你把他們放到一輛新車裏

and after a few laps, they've found the fastest line in that car,

跑幾個圈以後，他們就能夠為那新車找出最快的線道

and they're off to the races.

然後他們就準備好去比賽

It really makes you think,

你真的會覺得

we'd love to know what's going on inside their brain.

我們是多麼渴望知道他們的腦袋是如何運作的

So as researchers, that's what we decided to find out.

作為研究人員，這是我們決定追尋的問題

We decided to instrument not only the car,

我們不但決定裝備那輛車子

but also the race car driver,

還要裝備那個駕駛者

to try to get a glimpse into what was going on

去嘗試了解，在他們做出這些行為的時候

in their head as they were doing this.

他們的腦袋會產生怎麼的變化

Now, this is Dr. Lene Harbott applying electrodes

蘭尼•哈博 (Dr. Lene Harbott) 博士把電極

to the head of John Morton.

附在約翰•摩頓 (John Morton) 的頭上

John Morton is a former Can-Am and IMSA driver,

約翰•摩頓是加拿大－美國挑戰盃及國際賽車運動協會賽事的車手

who's also a class champion at Le Mans.

亦是利曼24小时耐力赛的分組冠軍

Fantastic driver, and very willing to put up with graduate students

這個優異的車手，而且非常願意和研究生合作

and this sort of research.

及參與這類研究

She's putting electrodes on his head

她正在把電極放到他的頭上

so that we can monitor the electrical activity

使我們能在約翰在賽道上飛馳的時候

in John's brain as he races around the track.

監察他的大腦活動

Now, clearly we're not going to put a couple of electrodes on his head

明顯地，我們不能以這幾支附在他頭上的電極

and understand exactly what all of his thoughts are on the track.

完全了解他在賽道的時候的思想

However, neuroscientists have identified certain patterns

但是，神經系統科學家已經認定了幾個不同的規律

that let us tease out some very important aspects of this.

為我們提供了幾個非常重要的線索

For instance, the resting brain

例如，大腦靜止的時候

tends to generate a lot of alpha waves.

傾向產生很多 alpha 波段

In contrast, theta waves are associated with

相對地，theta 波段與

a lot of cognitive activity, like visual processing,

很多認知活動有關，例如視覺訊息處理

things where the driver is thinking quite a bit.

這些都是駕駛者們經常想著的東西

Now, we can measure this,

現在，我們可以量度這些波段

and we can look at the relative power

以及我們可以看看

between the theta waves and the alpha waves.

theta 波段與alpha 波段的相對強弱

This gives us a measure of mental workload,

這會為我們提供一個大腦活動量的估算

how much the driver is actually challenged cognitively

看看駕駛者的大腦認知能力

at any point along the track.

受到多大的挑戰

Now, we wanted to see if we could actually record this

我們想看看我們能否在賽道上記錄這些數據

on the track, so we headed down south to Laguna Seca.

因此我們南下到加州的 Laguna Seca 賽車場

Laguna Seca is a legendary raceway

Laguna Seca 是一個傳奇的賽道

about halfway between Salinas and Monterey.

它在 Salinas 和 Monterey 兩個縣之間

It has a curve there called the Corkscrew.

它有一條彎彎的賽道叫做 "開瓶器"

Now, the Corkscrew is a chicane, followed by a quick

"開瓶器" 有幾個之字形的急彎

right-handed turn as the road drops three stories.

接著來一個急速的右轉，同時下墜約三層樓的高度

Now, the strategy for driving this as explained to me was,

有人告訴我處理這部分賽道的竅門

you aim for the bush in the distance,

你要望著遠處的矮樹

and as the road falls away, you realize it was actually the top of a tree.

當賽道下降的時候，你會發現那矮樹其實是一棵大樹的頂部

All right, so thanks to the Revs Program at Stanford,

好了，因為史丹福大學 Revs 計劃的資助

we were able to take John there

我們可以帶約翰到那裏

and put him behind the wheel

把他放到一輛1960年生產的

of a 1960 Porsche Abarth Carrera.

保時捷Abarth Carrera型號賽車的駕駛員座位上

Life is way too short for boring cars.

對於開無聊賽車，生命是如此短促！

So, here you see John on the track,

你可以看到約翰在賽道上

he's going up the hill -- Oh! Somebody liked that --

他正爬上那山丘-- 噢! 有人就是喜歡這樣--

and you can see, actually, his mental workload

你可以看到他的腦活動量

-- measuring here in the red bar --

-- 紅色那條軸 --

you can see his actions as he approaches.

你可以看到他上山時的動作

Now watch, he has to downshift.

現在又看! 他正下坡

And then he has to turn left.

接著他作了一個左轉

Look for the tree, and down.

看看那樹，再向下看

Not surprisingly, you can see this is a pretty challenging task.

可以預料的，你知道這是一個很具挑戰性的任務

You can see his mental workload spike as he goes through this,

你可以看到他跨越這裏的時候，腦活動量急劇增加

as you would expect with something that requires

對於一個這樣複雜的任務

this level of complexity.

我們覺得是合理的

But what's really interesting is to look at areas of the track

但有趣的地方是，你可以看到賽道上的某幾點

where his mental workload doesn't increase.

在這些地方約翰的腦活動量沒有增加

I'm going to take you around now

我會帶你往賽道

to the other side of the track.

的另一端看看

Turn three. And John's going to go into that corner

第三個彎: 約翰正準備穿越這個轉角

and the rear end of the car is going to begin to slide out.

賽車的尾輪開始滑行

He's going to have to correct for that with steering.

他要著手以方向盤改正這個情況

So watch as John does this here.

看看約翰怎樣做吧

Watch the mental workload, and watch the steering.

看看他的腦活動量，再看看那方向盤

The car begins to slide out, dramatic maneuver to correct it,

車子開始滑行，他以戲劇性的操作去改正它

and no change whatsoever in the mental workload.

但腦活動量並沒有改變

Not a challenging task.

這並非一個具挑戰性的任務

In fact, entirely reflexive.

實際上，這完全是個本能反應

Now, our data processing on this is still preliminary,

我們在這個情況所得的數據還在最初的階段

but it really seems that these phenomenal feats

但它好像說明了這些驚人的技藝背後

that the race car drivers are performing

賽車手的反應

are instinctive.

是出於本能的

They are things that they have simply learned to do.

這些動作都是他們學習的成果

It requires very little mental workload

對於他們要做出這些技藝

for them to perform these amazing feats.

這些動作對腦活動量的要求並不高

And their actions are fantastic.

他們的動作實在令人難以置信

This is exactly what you want to do on the steering wheel

這些都是每個人希望

to catch the car in this situation.

可以在這些情況下操作汽車的典範

Now, this has given us tremendous insight

這為我們提供了寶貴的參考價值

and inspiration for our own autonomous vehicles.

同時亦為我們自己的自動汽車提供相當多的啟發

We've started to ask the question:

我們開始問一個問題

Can we make them a little less algorithmic

我們能夠將自動汽車發展成較少依賴演算法

and a little more intuitive?

同時較多依賴直覺的機器嗎?

Can we take this reflexive action

我們可以把這樣的本能反應

that we see from the very best race car drivers,

這些我們可以在世界頂級賽車手的身上看到的本能反應

introduce it to our cars,

結合到我們的汽車上

and maybe even into a system that could

又或者結合到一個系統

get onto your car in the future?

在將來會放在你的汽車上的系統嗎?

That would take us a long step

要將這樣的系統放到汽車上

along the road to autonomous vehicles

還需要長時間完成的奮鬥

that drive as well as the best humans.

這個系統將可以與世上最好的車手比擬

But it's made us think a little bit more deeply as well.

但這亦引發出另一個比較深層次的問題

Do we want something more from our car

我們究竟想要一輛能做更多事情的汽車

than to simply be a chauffeur?

抑或只是一個司機呢?

Do we want our car to perhaps be a partner, a coach,

我們是否想汽車變成我們的伙伴、良師

someone that can use their understanding of the situation

一個會用它們對環境的理解

to help us reach our potential?

幫助我們發揮潛能的系統呢?

Can, in fact, the technology not simply replace humans,

究竟，科技是否可以不單單取代人類

but allow us to reach the level of reflex and intuition

而是誘發我們能力所及的

that we're all capable of?

本能反應及直覺呢?

So, as we move forward into this technological future,

因此，當我們在科技進步的同時

I want you to just pause and think of that for a moment.

我希望你停一停，思考這個問題

What is the ideal balance of human and machine?

甚麼是人類和機器最理想的平衡呢?

And as we think about that,

當我們思考這個問題的時候

let's take inspiration

就讓我們感受一下

from the absolutely amazing capabilities

人類身體及腦袋

of the human body and the human mind.

對我們的啟發吧!

Thank you.

謝謝大家

(Applause)

(掌聲)