Ken Goldberg: 4 lessons from robots about being human

Translator: Morton Bast Reviewer: Thu-Huong Ha

譯者: Ya-lun Lin 審譯者: James Hung

I know this is going to sound strange,

我知道這聽起來怪怪的

but I think robots can inspire us to be better humans.

但是我認為機器人能夠啟發我們

讓我們成為更好的人

我在美國賓州伯利恆(Bethlehem, Pennsylvania)長大

See, I grew up in Bethlehem, Pennsylvania,

the home of Bethlehem Steel.

伯利恆鋼鐵廠(Bethlehem Steel)就在那裡

My father was an engineer,

我父親是個工程師

在我成長的過程中

and when I was growing up, he would teach me how things worked.

他會敎了我工程的原理

我們會一起做一些東西

We would build projects together,

like model rockets and slot cars.

像是模型火箭和軌道車

Here's the go-kart that we built together.

這是我們一起做的卡丁車

站在輪胎後方的是我

That's me behind the wheel,

with my sister and my best friend at the time.

還有我的姐姐和我當時最好的朋友

And one day,

有一天我父親回到家

he came home, when I was about 10 years old,

當時我大概十歲

and at the dinner table, he announced

吃晚飯的時候

that for our next project, we were going to build ...

他宣布下一個我們要做的東西是機器人

a robot.

A robot.

機器人

我感到非常樂不可支

Now, I was thrilled about this,

because at school, there was a bully named Kevin,

因為學校裡

有一個愛欺負人的同學叫做凱文

他很愛找我麻煩

and he was picking on me,

because I was the only Jewish kid in class.

因為我是班上唯一的猶太人

So I couldn't wait to get started to work on this,

我等不及能夠快點開始做這件事情

so I could introduce Kevin to my robot.

這樣我就可以讓凱文「見識見識」我的機器人 (笑聲)

(Laughter)

(Robot noises)

(機器人的聲音)

(Laughter)

But that wasn't the kind of robot my dad had in mind.

這並不是當時我父親所想的那種機器人

(Laughter)

See, he owned a chromium-plating company,

當時他擁有一家鍍鉻公司

and they had to move heavy steel parts between tanks of chemicals.

他們得在一堆化學物品中

搬運沉重的鋼鐵零件

所以他需要一個像這樣的工業機器人

And so he needed an industrial robot like this,

that could basically do the heavy lifting.

來搬運重物

But my dad didn't get the kind of robot he wanted, either.

但是我父親並沒有得到他想要的那種機器人

He and I worked on it for several years,

我父親和我在這件事情上努力了好幾年

but it was the 1970s, and the technology that was available to amateurs

但是當時是1970年代

還沒有出現

非專業人士能夠取得的技術

just wasn't there yet.

So Dad continued to do this kind of work by hand.

因此我父親靠著人工繼續這類工作

And a few years later,

幾年後

he was diagnosed with cancer.

他被診斷患有癌症

You see,

你看，當時我們想做的那種機器人

what the robot we were trying to build was telling him

不是在告訴他不要搬重物

was not about doing the heavy lifting.

It was a warning

而是對他暴露在有毒的化學物品中的一種警告

about his exposure to the toxic chemicals.

當時他沒有意會到

He didn't recognize that at the time,

and he contracted leukemia.

所以得了血癌

And he died at the age of 45.

在45歲的時候過世

這件事情讓我受到很大的打擊

I was devastated by this.

And I never forgot the robot that he and I tried to build.

我永遠忘不了我父親和我想做的那種機器人

When I was at college, I decided to study engineering, like him.

大學的時候，我決定像父親一樣主修工程

我進了卡內基美隆大學，取得機器人學博士學位

And I went to Carnegie Mellon, and I earned my PhD in robotics.

I've been studying robots ever since.

在那之後，我一直在研究機器人

So what I'd like to tell you about are four robot projects,

我想向你們介紹的是

四項機器人設計

以及這些設計如何啟發我成為更好的人

and how they've inspired me to be a better human.

By 1993, I was a young professor at USC,

1993年以前 我是南加大(USC)裡的一個年輕教授

and I was just building up my own robotics lab,

當時才要開始建立我個人的機器人實驗室

and this was the year the World Wide Web came out.

網路的出現也是在這一年

And I remember my students were the ones who told me about it,

我記得是我的學生

告訴我網路的出現

and we would -- we were just amazed.

我們對此讚嘆不已

We started playing with this, and that afternoon,

我們開始摸索網路

we realized that we could use this new, universal interface

某個下午 我們發現可以利用這個縱橫全球的平台

to allow anyone in the world to operate the robot in our lab.

讓全世界任何一個人

來操作我們實驗室中的機器人

我們捨棄將它做為戰鬥或工業使用

So, rather than have it fight or do industrial work,

we decided to build a planter,

選擇用它來耕種

put the robot into the center of it,

我們把機器人放在這東西的中心位置

and we called it the Telegarden.

並稱它為天線花園

And we had put a camera in the gripper of the hand of the robot,

我們在機器人手上的夾具裝設了照相機

也寫了專屬的程式和軟體

and we wrote some special scripts and software,

如此一來

so that anyone in the world could come in,

世界各地的人就可以透過點擊螢幕

and by clicking on the screen,

they could move the robot around and visit the garden.

進來任意移動機器人

並且參觀花園

But we also set up some other software

我們同時也設有其他軟體

that lets you participate and help us water the garden, remotely.

讓人們進來幫忙我們遙控澆水

在你澆過幾次水之後

And if you watered it a few times,

we'd give you your own seed to plant.

我們就會給你種子去栽種

Now, this was an engineering project,

這是一個設計，一個工程設計

and we published some papers on the system design of it,

我們出版了一些

天線花園系統設計的文件

but we also thought of it as an art installation.

但我們也視天線花園為一種裝置藝術

It was invited, after the first year,

天線花園出現一年後

by the Ars Electronica Museum in Austria,

受到了奧地利的電子藝術博物館邀請

to have it installed in their lobby.

將其設置在博物館大廳

並且很榮幸地在博物館網站上

And I'm happy to say, it remained online there, 24 hours a day,

每天24小時展出將近九年時間

for almost nine years.

That robot was operated by more people

操作這個機器人的人數

than any other robot in history.

比歷史上任何其他機器人都還要多

Now, one day,

有一天

I got a call out of the blue from a student,

我突然接到

一個學生的電話

who asked a very simple but profound question.

他問了我一個非常簡單卻意義深遠的問題

他說，天線花園是真的嗎?

He said, "Is the robot real?"

Now, everyone else had assumed it was,

當時所有人都認為那是真的

and we knew it was, because we were working with it.

而我們是製作人所以我們知道

But I knew what he meant,

但是我能理解他為什麼這麼問

because it would be possible

因為那有可能是拍下一堆花朵在花園裡的照片

to take a bunch of pictures of flowers in a garden

然後利用電腦系統合成出來的

and then, basically, index them in a computer system,

看起來像是有一個真的機器人在那裡

such that it would appear that there was a real robot,

但實際上並不存在

when there wasn't.

And the more I thought about it,

我想了很久

I couldn't think of a good answer for how he could tell the difference.

卻想不出能讓他分辨兩者差異的方法

這件事情差不多是發生在

This was right about the time that I was offered a position

柏克萊大學(Berkeley)請我過來工作的時候

here at Berkeley.

And when I got here,

我到達這裡時

I looked up Hubert Dreyfus,

who's a world-renowned professor of philosophy,

拜訪了世界著名的哲學家德雷佛斯

And I talked with him about this and he said,

我和他談到這件事情，他說

這是最古老也最重要的哲學問題之一

"This is one of the oldest and most central problems in philosophy.

這要回顧到懷疑主義者

It goes back to the Skeptics and up through Descartes.

並且向上追溯至笛卡兒

It's the issue of epistemology,

這是認識知識論的問題

the study of how do we know that something is true."

研究如何判別事物真假

So he and I started working together,

我們開始一起研究

and we coined a new term: "telepistemology,"

創造出一個新名詞：天線認識論

是對遠距知識的研究

the study of knowledge at a distance.

We invited leading artists, engineers and philosophers

我們邀請了藝術家、工程師以及哲學家的先驅

來寫一篇相關報告

to write essays about this,

and the results are collected in this book from MIT Press.

研究結果發表在麻省理工學院

出版的這本書裡

So thanks to this student,

我要感謝這名提出疑問的學生

who questioned what everyone else had assumed to be true,

他質疑了所有人都相信的事實

this project taught me an important lesson about life,

這件事情給我上了人生的一課

which is to always question assumptions.

就是我們要對任何假設提出質疑

Now, the second project I'll tell you about

我要講的第二項設計

grew out of the Telegarden.

是在天線花園之後發展出來的

As it was operating, my students and I were very interested

天線花園還在運作的時候

in how people were interacting with each other,

我的學生和我對於人與人的互動

and what they were doing with the garden.

以及他們在花園中所做的事情很感興趣

我們開始思考

So we started thinking:

what if the robot could leave the garden

如果機器人可以離開花園

and go out into some other interesting environment?

進入其他有意思的環境中會怎麼樣呢?

Like, for example, what if it could go to a dinner party

比方說

在白宮舉行的晚宴? (笑聲)

at the White House?

(Laughter)

So, because we were interested more in the system design

比起硬體設備

and the user interface than in the hardware,

我們對系統設計和使用者平台更感興趣

we decided that,

因此我們決定

rather than have a robot replace the human to go to the party,

讓人來取代機器人

we'd have a human replace the robot.

而不是機器人取代人去參加宴會

We called it the Tele-Actor.

我們稱之為人形機器人

We got a human,

我們找到一個人

someone who's very outgoing and gregarious,

她非常外向又熱衷社交

and she was outfitted with a helmet with various equipment,

她戴著一頂頭盔

上面有各種設備、攝影機和麥克風

cameras and microphones,

and then a backpack with wireless Internet connection.

背著一個備有無線網路連線功能的背包

And the idea was that she could go

我們的想法是

into a remote and interesting environment,

她可以進入一個有意思的遠距環境

and then over the Internet,

people could experience what she was experiencing.

人們可以透過網路體驗她所經歷的

So they could see what she was seeing,

看到她所看的

but then, more importantly, they could participate,

更重要的是

by interacting with each other and coming up with ideas

人們可以透過與其他人的互動參與其中

關於她該做什麼去哪裡

about what she should do next and where she should go,

我們想了很多點子

and then conveying those to the Tele-Actor.

並且將這些都傳輸到人形機器人上

So we got a chance to take the Tele-Actor

我們得到機會，把人形機器人

to the Webby Awards in San Francisco.

帶到在舊金山舉行的威比獎典禮上

And that year, Sam Donaldson was the host.

那年的主持人是山姆．唐納森

在幕簾拉開之前

Just before the curtain went up, I had about 30 seconds

我有大概30秒時間向唐納森先生說明我們要做的事

to explain to Mr. Donaldson what we were going to do.

我說

And I said, "The Tele-Actor is going to be joining you onstage.

人形機器人會和你一起出現在舞台上

This is a new experimental project,

這是一項實驗性的新設計

人們會從自己的螢幕上看到她

and people are watching her on their screens,

there's cameras involved and there's microphones

她身上有攝影機和麥克風

單邊耳朵會戴耳機

and she's got an earbud in her ear,

and people over the network are giving her advice

她下一步該怎麼做

人們會透過網路給她出主意

about what to do next."

唐納森先生說

And he said, "Wait a second.

That's what I do."

慢著，那是我的工作 (笑聲)

(Laughter)

So he loved the concept,

他很喜歡這個概念

and when the Tele-Actor walked onstage, she walked right up to him,

在人形機器人步上舞台時

她直接朝唐納森先生走去

and she gave him a big kiss right on the lips.

並且在他嘴唇上用力親了一下 (笑聲)

(Laughter)

We were totally surprised -- we had no idea that would happen.

我們都非常驚訝

這是我們意料之外的事情

And he was great, he just gave her a big hug in return,

唐納森先生人非常好，回給她一個大擁抱

and it worked out great.

得到熱烈的回響

But that night, as we were packing up,

當天晚上，在我們打包的時候

I asked the Tele-Actor, how did the Tele-Directors decide

我問了人形機器人

導演是如何決定讓她親山姆‧唐納森的

that they would give a kiss to Sam Donaldson?

And she said they hadn't.

她說他們並沒有要她這麼做

She said, when she was just about to walk onstage,

她說在她上台之前

the Tele-Directors still were trying to agree on what to do,

導演們還在討論要怎麼做

and so she just walked onstage and did what felt most natural.

所以她就走上台

做了她覺得應該做的(笑聲)

(Laughter)

So, the success of the Tele-Actor that night

因此，那晚人形機器人的成功

was due to the fact that she was a wonderful actor.

完全歸功於她是個很棒的演員

She knew when to trust her instincts.

她知道什麼時候該相信自己的直覺

And so that project taught me another lesson about life,

這個設計給我上了人生中另一課

which is that, when in doubt, improvise.

不確定怎麼做的時候，要隨機應變 (笑聲)

(Laughter)

Now, the third project grew out of my experience

第三項設計是源自我父親住院時

我個人的經驗

when my father was in the hospital.

當時我父親在接受化療

He was undergoing a treatment -- chemotherapy treatments --

還有另一種相關療法

and there's a related treatment called brachytherapy,

稱為近距放射療法

where tiny, radioactive seeds are placed into the body

是將微小的放射性種子植入體內來治療癌症腫瘤

to treat cancerous tumors.

And the way it's done, as you can see here,

這個療法正如你所見

is that surgeons insert needles into the body

外科醫師會把針插在人體上

to deliver the seeds.

來傳遞放射性種子

And all these needles are inserted in parallel.

這些插入人體的針呈平行排列

So it's very common that some of the needles penetrate sensitive organs.

因此有些針

很容易會刺入敏感的器官

And as a result, the needles damage these organs, cause damage,

結果這些針破壞了器官

which leads to trauma and side effects.

引起器官損傷及副作用

So my students and I wondered:

我的學生和我開始想

what if we could modify the system,

我們是不是可以改善這個系統

so that the needles could come in at different angles?

讓針可以從不同角度插入

So we simulated this;

我們做了模擬

we developed some optimization algorithms and we simulated this.

找出最佳的運算法則並再次模擬

And we were able to show

我們有辦法做到避開脆弱的器官

that we are able to avoid the delicate organs,

同時仍然能夠維持

and yet still achieve the coverage of the tumors with the radiation.

放射線在腫瘤上的覆蓋範圍

So now, we're working with doctors at UCSF

我們現在和加州大學舊金山分校裡的醫生

and engineers at Johns Hopkins,

以及約翰霍普金斯大學裡的工程師合作

and we're building a robot that has a number of --

我們正在做一種機器人

it's a specialized design with different joints

它經過特殊設計，有許多的關節

that can allow the needles to come in at an infinite variety of angles.

讓針能夠以各種不同角度插入

And as you can see here, they can avoid delicate organs

如你所見，他們可以避開脆弱的器官

在需要的地方插針

and still reach the targets they're aiming for.

透過對針必須以平行排列有所懷疑

So, by questioning this assumption that all the needles have to be parallel,

我又在這個設計上學到一件事情

this project also taught me an important lesson:

就是當我們心中有疑問時

When in doubt, when your path is blocked, pivot.

當你眼前的路受到阻斷時，就換條路走吧 (投影片翻譯) 1. 要質疑假設 2. 猶疑不定時，要隨機應變 3. 此路不通時，要繞道而行

And the last project also has to do with medical robotics.

最後一個設計也是跟醫學機器人有關

And this is something that's grown out of a system

這是從一個叫做

called the da Vinci surgical robot.

達文西手術機器人的系統中發展出來的

And this is a commercially available device.

這個設備市面上可以買得到

It's being used in over 2,000 hospitals around the world.

全球已經有超過兩千家醫院使用這項設備

這個創意讓外科醫生

The idea is it allows the surgeon to operate comfortably

能夠在屬於自己的空間裡自在地做手術

in his own coordinate frame.

但是許多手術的附加工作

Many of the subtasks in surgery are very routine and tedious, like suturing,

比方說縫合，是非常一成不變又沉悶的

and currently, all of these are performed

目前這些工作的執行

under the specific and immediate control of the surgeon.

都是直接由外科醫生專門管控

So the surgeon becomes fatigued over time.

造成外科醫生的過度疲勞

我們一直在思考

And we've been wondering,

what if we could program the robot to perform some of these subtasks,

如果我們能設計讓機器人

來執行部分的附加工作

and thereby free the surgeon

就可以讓外科醫生

to focus on the more complicated parts of the surgery,

專心投入在手術中更複雜的部分

and also cut down on the time that the surgery would take

如果我們能夠讓機器人的動作加快一些

if we could get the robot to do them a little bit faster?

也能夠縮短手術時間

Now, it's hard to program a robot to do delicate things like this.

要設計讓機器人做這種精細的工作是很困難的

但是我柏克萊大學的同事

But it turns out my colleague Pieter Abbeel, who's here at Berkeley,

皮耶特‧阿布比爾做到了

has developed a new set of techniques for teaching robots from example.

他研發出一套讓機器人從實例中學習的新技術

So he's gotten robots to fly helicopters,

透過觀摩專家的飛行

do incredibly interesting, beautiful acrobatics,

他成功讓機器人駕駛直升機

by watching human experts fly them.

做一些有意思又讓人驚艷的特技

我們取得一個這種機器人

So we got one of these robots.

We started working with Pieter and his students.

開始和皮耶特還有他的學生一起研究

And we asked a surgeon to perform a task --

我們請一個外科醫生執行手術

我們的做法是和機器人一起

with the robot.

So what we're doing is asking the surgeon to perform the task,

我們要求機器人

也進行手術

and we record the motions of the robot.

然後我們把機器人的一舉一動都錄下來

So here's an example.

接下來我用八字縫合

I'll use tracing out a figure eight as an example.

做為例子說明

這裡有三個例子

So here's what it looks like when the robot --

this is what the robot's path looks like, those three examples.

是由機器人所做的

八字縫合

Now, those are much better than what a novice like me could do,

這遠比我這個外行人所能做的要強得多

但是還是不夠純熟精準

but they're still jerky and imprecise.

So we record all these examples, the data,

所以我們記錄所有的例子和數據

然後仔細看過一連串的步驟

and then go through a sequence of steps.

首先，我們使用了一個取自語言識別

First, we use a technique called dynamic time warping

from speech recognition.

名為動態時間校正的技術

And this allows us to temporally align all of the examples.

這讓我們能夠暫時結合所有的實例

And then we apply Kalman filtering, a technique from control theory,

接著我們採用卡爾曼濾波器

這是一門採自控制理論的技術

that allows us to statistically analyze all the noise

這個技術讓我們可以對所有雜訊做數據分析

節取我們想要的軌跡

and extract the desired trajectory that underlies them.

Now we take those human demonstrations --

我們的做法是

they're all noisy and imperfect --

從這些嘈雜又不盡完美的人體示範中

and we extract from them an inferred task trajectory

取出套用在機器人身上的

and control sequence for the robot.

任務判別軌跡以及控制程式

We then execute that on the robot,

接著我們在機器人上面執行

we observe what happens,

觀察整個過程

then we adjust the controls,

然後利用稱為反覆學習的一系列技術

using a sequence of techniques called iterative learning.

來調整控制器

Then what we do is we increase the velocity a little bit.

接著我們做的是，稍微調快速率

We observe the results, adjust the controls again,

我們觀察結果，又調整了控制器

再接著觀察

and observe what happens.

And we go through this several rounds.

我們重覆這個過程好幾次

And here's the result.

得到這個結果

That's the inferred task trajectory,

這個是推論的工作軌跡

and here's the robot moving at the speed of the human.

這個則是機器人與人類以同樣速度移動的軌跡

Here's four times the speed of the human.

這是比人類快四倍的軌跡

Here's seven times.

這是快七倍的

And here's the robot operating at 10 times the speed of the human.

這是機器人以比人類快十倍的速度

操作的軌跡

So we're able to get a robot to perform a delicate task

所以說，我們是有辦法讓機器人

like a surgical subtask,

用比人類快十倍的速度

at 10 times the speed of a human.

來執行像手術附加工作這樣的精細工作的

So this project also,

因為這個設計包含了練習和學習

because of its involved practicing and learning,

多次重覆做同一件事情

doing something over and over again,

this project also has a lesson, which is:

所以這個設計也教了我們一件事情

if you want to do something well,

就是如果你想做好一件事情

就只能不斷地練習再練習

there's no substitute for practice, practice, practice.

這些就是這幾年來

So these are four of the lessons that I've learned from robots

我從機器人身上學到的四門課

over the years.

至於機器人學

And the field of robotics has gotten much better over time.

這個領域也隨著時間進步了許多

Nowadays, high school students can build robots,

如今就連高中學生

like the industrial robot my dad and I tried to build.

也能做出像我和我父親嘗試做的那種工業機器人

But, it's very -- now ...

現在我有個女兒

And now, I have a daughter,

named Odessa.

名字叫歐妲莎(Odessa)

She's eight years old.

她現在八歲

And she likes robots, too.

她也喜歡機器人

Maybe it runs in the family.

這大概是家族風氣(笑聲)

(Laughter)

I wish she could meet my dad.

我希望她有機會見到我父親

And now I get to teach her how things work,

現在換我教她做事情的方法

and we get to build projects together.

我們一起做很多東西

And I wonder what kind of lessons she'll learn from them.

我常常想她能夠從中學到什麼

Robots are the most human of our machines.

機器人是所有機器中

最具有人性的

They can't solve all of the world's problems,

它們無法解決世界上的所有問題

but I think they have something important to teach us.

但我認為我們可以從它們身上學到一些重要的事

I invite all of you

我希望各位去思考

to think about the innovations that you're interested in,

那些你感興趣的發明

the machines that you wish for.

你夢想中的機器

And think about what they might be telling you.

想想這些東西可能要傳達的訊息

因為我的直覺告訴我

Because I have a hunch that many of our technological innovations,

許多我們夢想中的

the devices we dream about,

科技發明和設備

可以啟發我們成為更好的人

can inspire us to be better humans.

Thank you.

謝謝(掌聲)

(Applause)