Placeholder Image

Subtitles section Play video

  • 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

  • See, I grew up in Bethlehem, Pennsylvania,

    讓我們成為更好的人

  • the home of Bethlehem Steel.

    我在美國賓州伯利恆(Bethlehem, Pennsylvania)長大

  • My father was an engineer,

    伯利恆鋼鐵廠(Bethlehem Steel)就在那裡

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

    我父親是個工程師

  • We would build projects together,

    在我成長的過程中

  • like model rockets and slot cars.

    他會敎了我工程的原理

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

    我們會一起做一些東西

  • That's me behind the wheel,

    像是模型火箭和軌道車

  • with my sister and my best friend at the time.

    這是我們一起做的卡丁車

  • And one day,

    站在輪胎後方的是我

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

    還有我的姐姐和我當時最好的朋友

  • and at the dinner table, he announced

    有一天我父親回到家

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

    當時我大概十歲

  • a robot.

    吃晚飯的時候

  • A robot.

    他宣布下一個我們要做的東西是機器人

  • Now, I was thrilled about this,

    機器人

  • because at school, there was a bully named Kevin,

    我感到非常樂不可支

  • and he was picking on me,

    因為學校裡

  • because I was the only Jewish kid in class.

    有一個愛欺負人的同學叫做凱文

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

    他很愛找我麻煩

  • so I could introduce Kevin to my robot.

    因為我是班上唯一的猶太人

  • (Laughter)

    我等不及能夠快點開始做這件事情

  • (Robot noises)

    這樣我就可以讓凱文「見識見識」我的機器人 (笑聲)

  • (Laughter)

    (機器人的聲音)

  • 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)

    當時他擁有一家鍍鉻公司

  • See, he owned a chromium-plating company,

    他們得在一堆化學物品中

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

    搬運沉重的鋼鐵零件

  • And so he needed an industrial robot like this,

    所以他需要一個像這樣的工業機器人

  • 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

  • just wasn't there yet.

    還沒有出現

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

    非專業人士能夠取得的技術

  • And a few years later,

    因此我父親靠著人工繼續這類工作

  • he was diagnosed with cancer.

    幾年後

  • You see,

    他被診斷患有癌症

  • what the robot we were trying to build was telling him

    你看,當時我們想做的那種機器人

  • was not about doing the heavy lifting.

    不是在告訴他不要搬重物

  • 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,

    所以得了血癌

  • and he contracted leukemia.

    在45歲的時候過世

  • And he died at the age of 45.

    這件事情讓我受到很大的打擊

  • I was devastated by this.

    我永遠忘不了我父親和我想做的那種機器人

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

    大學的時候,我決定像父親一樣主修工程

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

    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.

    在那之後,我一直在研究機器人

  • I've been studying robots ever since.

    我想向你們介紹的是

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

    四項機器人設計

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

    以及這些設計如何啟發我成為更好的人

  • 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

  • and we would -- we were just amazed.

    告訴我網路的出現

  • We started playing with this, and that afternoon,

    我們對此讚嘆不已

  • we realized that we could use this new, universal interface

    我們開始摸索網路

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

    某個下午 我們發現可以利用這個縱橫全球的平台

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

    讓全世界任何一個人

  • we decided to build a planter,

    來操作我們實驗室中的機器人

  • put the robot into the center of it,

    我們捨棄將它做為戰鬥或工業使用

  • and we called it the Telegarden.

    選擇用它來耕種

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

    我們把機器人放在這東西的中心位置

  • and we wrote some special scripts and software,

    並稱它為天線花園

  • so that anyone in the world could come in,

    我們在機器人手上的夾具裝設了照相機

  • and by clicking on the screen,

    也寫了專屬的程式和軟體

  • they could move the robot around and visit the garden.

    如此一來

  • But we also set up some other software

    世界各地的人就可以透過點擊螢幕

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

    進來任意移動機器人

  • And if you watered it a few times,

    並且參觀花園

  • we'd give you your own seed to plant.

    我們同時也設有其他軟體

  • Now, this was an engineering project,

    讓人們進來幫忙我們遙控澆水

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

    在你澆過幾次水之後

  • but we also thought of it as an art installation.

    我們就會給你種子去栽種

  • It was invited, after the first year,

    這是一個設計,一個工程設計

  • by the Ars Electronica Museum in Austria,

    我們出版了一些

  • to have it installed in their lobby.

    天線花園系統設計的文件

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

    但我們也視天線花園為一種裝置藝術

  • for almost nine years.

    天線花園出現一年後

  • That robot was operated by more people

    受到了奧地利的電子藝術博物館邀請

  • than any other robot in history.

    將其設置在博物館大廳

  • Now, one day,

    並且很榮幸地在博物館網站上

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

    每天24小時展出將近九年時間

  • who asked a very simple but profound question.

    操作這個機器人的人數

  • He said, "Is the robot real?"

    比歷史上任何其他機器人都還要多

  • Now, everyone else had assumed it was,

    有一天

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

    我突然接到

  • But I knew what he meant,

    一個學生的電話

  • because it would be possible

    他問了我一個非常簡單卻意義深遠的問題

  • to take a bunch of pictures of flowers in a garden

    他說,天線花園是真的嗎?

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

    當時所有人都認為那是真的

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

    而我們是製作人所以我們知道

  • when there wasn't.

    但是我能理解他為什麼這麼問

  • And the more I thought about it,

    因為那有可能是拍下一堆花朵在花園裡的照片

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

    然後利用電腦系統合成出來的

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

    看起來像是有一個真的機器人在那裡

  • here at Berkeley.

    但實際上並不存在

  • And when I got here,

    我想了很久

  • I looked up Hubert Dreyfus,

    卻想不出能讓他分辨兩者差異的方法

  • who's a world-renowned professor of philosophy,

    這件事情差不多是發生在

  • And I talked with him about this and he said,

    柏克萊大學(Berkeley)請我過來工作的時候

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

    我到達這裡時

  • It goes back to the Skeptics and up through Descartes.

    拜訪了世界著名的哲學家德雷佛斯

  • It's the issue of epistemology,

    我和他談到這件事情,他說

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

    這是最古老也最重要的哲學問題之一

  • So he and I started working together,

    這要回顧到懷疑主義者

  • and we coined a new term: "telepistemology,"

    並且向上追溯至笛卡兒

  • the study of knowledge at a distance.

    這是認識知識論的問題

  • We invited leading artists, engineers and philosophers

    研究如何判別事物真假

  • to write essays about this,

    我們開始一起研究

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

    創造出一個新名詞:天線認識論

  • So thanks to this student,

    是對遠距知識的研究

  • who questioned what everyone else had assumed to be true,

    我們邀請了藝術家、工程師以及哲學家的先驅

  • this project taught me an important lesson about life,

    來寫一篇相關報告

  • which is to always question assumptions.

    研究結果發表在麻省理工學院

  • Now, the second project I'll tell you about

    出版的這本書裡

  • grew out of the Telegarden.

    我要感謝這名提出疑問的學生

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

    他質疑了所有人都相信的事實

  • in how people were interacting with each other,

    這件事情給我上了人生的一課

  • and what they were doing with the garden.

    就是我們要對任何假設提出質疑

  • So we started thinking:

    我要講的第二項設計

  • what if the robot could leave the garden

    是在天線花園之後發展出來的

  • and go out into some other interesting environment?

    天線花園還在運作的時候

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

    我的學生和我對於人與人的互動

  • at the White House?

    以及他們在花園中所做的事情很感興趣

  • (Laughter)

    我們開始思考

  • So, because we were interested more in the system design

    如果機器人可以離開花園

  • and the user interface than in the hardware,

    進入其他有意思的環境中會怎麼樣呢?

  • we decided that,

    比方說

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

    在白宮舉行的晚宴? (笑聲)

  • we'd have a human replace the robot.

    比起硬體設備

  • We called it the Tele-Actor.

    我們對系統設計和使用者平台更感興趣

  • We got a human,

    因此我們決定

  • someone who's very outgoing and gregarious,

    讓人來取代機器人

  • and she was outfitted with a helmet with various equipment,

    而不是機器人取代人去參加宴會

  • cameras and microphones,

    我們稱之為人形機器人

  • and then a backpack with wireless Internet connection.

    我們找到一個人

  • And the idea was that she could go

    她非常外向又熱衷社交

  • into a remote and interesting environment,

    她戴著一頂頭盔

  • and then over the Internet,

    上面有各種設備、攝影機和麥克風

  • people could experience what she was experiencing.

    背著一個備有無線網路連線功能的背包

  • So they could see what she was seeing,

    我們的想法是

  • but then, more importantly, they could participate,

    她可以進入一個有意思的遠距環境

  • by interacting with each other and coming up with ideas

    人們可以透過網路體驗她所經歷的

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

    看到她所看的

  • and then conveying those to the Tele-Actor.

    更重要的是

  • So we got a chance to take the Tele-Actor

    人們可以透過與其他人的互動參與其中

  • to the Webby Awards in San Francisco.

    關於她該做什麼去哪裡

  • And that year, Sam Donaldson was the host.

    我們想了很多點子

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

    並且將這些都傳輸到人形機器人上

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

    我們得到機會,把人形機器人

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

    帶到在舊金山舉行的威比獎典禮上

  • This is a new experimental project,

    那年的主持人是山姆.唐納森

  • and people are watching her on their screens,

    在幕簾拉開之前

  • there's cameras involved and there's microphones

    我有大概30秒時間向唐納森先生說明我們要做的事

  • and she's got an earbud in her ear,

    我說

  • and people over the network are giving her advice

    人形機器人會和你一起出現在舞台上

  • about what to do next."

    這是一項實驗性的新設計

  • And he said, "Wait a second.

    人們會從自己的螢幕上看到她

  • That's what I do."

    她身上有攝影機和麥克風

  • (Laughter)

    單邊耳朵會戴耳機

  • So he loved the concept,

    她下一步該怎麼做

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

    人們會透過網路給她出主意

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

    唐納森先生說

  • (Laughter)

    慢著,那是我的工作 (笑聲)

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

    他很喜歡這個概念

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

    在人形機器人步上舞台時

  • and it worked out great.

    她直接朝唐納森先生走去

  • But that night, as we were packing up,

    並且在他嘴唇上用力親了一下 (笑聲)

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

    我們都非常驚訝

  • that they would give a kiss to Sam Donaldson?

    這是我們意料之外的事情

  • And she said they hadn't.

    唐納森先生人非常好,回給她一個大擁抱

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

    得到熱烈的回響

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

    當天晚上,在我們打包的時候

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

    我問了人形機器人

  • (Laughter)

    導演是如何決定讓她親山姆‧唐納森的

  • So, the success of the Tele-Actor that night

    她說他們並沒有要她這麼做

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

    她說在她上台之前

  • She knew when to trust her instincts.

    導演們還在討論要怎麼做

  • And so that project taught me another lesson about life,

    所以她就走上台

  • which is that, when in doubt, improvise.

    做了她覺得應該做的(笑聲)

  • (Laughter)

    因此,那晚人形機器人的成功

  • Now, the third project grew out of my experience

    完全歸功於她是個很棒的演員

  • when my father was in the hospital.

    她知道什麼時候該相信自己的直覺

  • He was undergoing a treatment -- chemotherapy treatments --

    這個設計給我上了人生中另一課

  • and there's a related treatment called brachytherapy,

    不確定怎麼做的時候,要隨機應變 (笑聲)

  • where tiny, radioactive seeds are placed into the body

    第三項設計是源自我父親住院時

  • to treat cancerous tumors.

    我個人的經驗

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

    當時我父親在接受化療

  • is that surgeons insert needles into the body

    還有另一種相關療法

  • to deliver the seeds.

    稱為近距放射療法

  • And all these needles are inserted in parallel.

    是將微小的放射性種子植入體內來治療癌症腫瘤

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

    這個療法正如你所見

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

    外科醫師會把針插在人體上

  • which leads to trauma and side effects.

    來傳遞放射性種子

  • So my students and I wondered:

    這些插入人體的針呈平行排列

  • what if we could modify the system,

    因此有些針

  • so that the needles could come in at different angles?

    很容易會刺入敏感的器官

  • So we simulated this;

    結果這些針破壞了器官

  • we developed some optimization algorithms and we simulated this.

    引起器官損傷及副作用

  • And we were able to show

    我的學生和我開始想

  • that we are able to avoid the delicate organs,

    我們是不是可以改善這個系統

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

    讓針可以從不同角度插入

  • So now, we're working with doctors at UCSF

    我們做了模擬

  • and engineers at Johns Hopkins,

    找出最佳的運算法則並再次模擬

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

    我們有辦法做到避開脆弱的器官

  • it's a specialized design with different joints

    同時仍然能夠維持

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

    放射線在腫瘤上的覆蓋範圍

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

    我們現在和加州大學舊金山分校裡的醫生

  • and still reach the targets they're aiming for.

    以及約翰霍普金斯大學裡的工程師合作

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

    我們正在做一種機器人

  • this project also taught me an important lesson:

    它經過特殊設計,有許多的關節

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

    讓針能夠以各種不同角度插入

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

    如你所見,他們可以避開脆弱的器官

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

    在需要的地方插針

  • called the da Vinci surgical robot.

    透過對針必須以平行排列有所懷疑

  • And this is a commercially available device.

    我又在這個設計上學到一件事情

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

    就是當我們心中有疑問時

  • The idea is it allows the surgeon to operate comfortably

    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

  • in his own coordinate frame.

    最後一個設計也是跟醫學機器人有關

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

    這是從一個叫做

  • and currently, all of these are performed

    達文西手術機器人的系統中發展出來的

  • under the specific and immediate control of the surgeon.

    這個設備市面上可以買得到

  • So the surgeon becomes fatigued over time.

    全球已經有超過兩千家醫院使用這項設備

  • And we've been wondering,

    這個創意讓外科醫生

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

    能夠在屬於自己的空間裡自在地做手術

  • and thereby free the surgeon

    但是許多手術的附加工作

  • to focus on the more complicated parts of the surgery,

    比方說縫合,是非常一成不變又沉悶的

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

    目前這些工作的執行

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

    都是直接由外科醫生專門管控

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

    造成外科醫生的過度疲勞

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

    我們一直在思考

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

    如果我們能設計讓機器人

  • So he's gotten robots to fly helicopters,

    來執行部分的附加工作

  • do incredibly interesting, beautiful acrobatics,

    就可以讓外科醫生

  • by watching human experts fly them.

    專心投入在手術中更複雜的部分

  • So we got one of these robots.

    如果我們能夠讓機器人的動作加快一些

  • We started working with Pieter and his students.

    也能夠縮短手術時間

  • And we asked a surgeon to perform a task --

    要設計讓機器人做這種精細的工作是很困難的

  • with the robot.

    但是我柏克萊大學的同事

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

    皮耶特‧阿布比爾做到了

  • and we record the motions of the robot.

    他研發出一套讓機器人從實例中學習的新技術

  • So here's an example.

    透過觀摩專家的飛行

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

    他成功讓機器人駕駛直升機

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

    做一些有意思又讓人驚艷的特技

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

    我們取得一個這種機器人

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

    開始和皮耶特還有他的學生一起研究

  • but they're still jerky and imprecise.

    我們請一個外科醫生執行手術

  • So we record all these examples, the data,

    我們的做法是和機器人一起

  • and then go through a sequence of steps.

    我們要求機器人

  • First, we use a technique called dynamic time warping

    也進行手術

  • from speech recognition.

    然後我們把機器人的一舉一動都錄下來

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

    接下來我用八字縫合

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

    做為例子說明

  • that allows us to statistically analyze all the noise

    這裡有三個例子

  • and extract the desired trajectory that underlies them.

    是由機器人所做的

  • Now we take those human demonstrations --

    八字縫合

  • they're all noisy and imperfect --

    這遠比我這個外行人所能做的要強得多

  • and we extract from them an inferred task trajectory

    但是還是不夠純熟精準

  • and control sequence for the robot.

    所以我們記錄所有的例子和數據

  • We then execute that on the robot,

    然後仔細看過一連串的步驟

  • we observe what happens,

    首先,我們使用了一個取自語言識別

  • then we adjust the controls,

    名為動態時間校正的技術

  • using a sequence of techniques called iterative learning.

    這讓我們能夠暫時結合所有的實例

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

    接著我們採用卡爾曼濾波器

  • We observe the results, adjust the controls again,

    這是一門採自控制理論的技術

  • and observe what happens.

    這個技術讓我們可以對所有雜訊做數據分析

  • And we go through this several rounds.

    節取我們想要的軌跡

  • And here's the result.

    我們的做法是

  • That's the inferred task trajectory,

    從這些嘈雜又不盡完美的人體示範中

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

    取出套用在機器人身上的

  • Here's four times the speed of the human.

    任務判別軌跡以及控制程式

  • Here's seven times.

    接著我們在機器人上面執行

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

    觀察整個過程

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

    然後利用稱為反覆學習的一系列技術

  • like a surgical subtask,

    來調整控制器

  • at 10 times the speed of a human.

    接著我們做的是,稍微調快速率

  • So this project also,

    我們觀察結果,又調整了控制器

  • because of its involved practicing and learning,

    再接著觀察

  • doing something over and over again,

    我們重覆這個過程好幾次

  • this project also has a lesson, which is:

    得到這個結果

  • if you want to do something well,

    這個是推論的工作軌跡

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

    這個則是機器人與人類以同樣速度移動的軌跡

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

    這是比人類快四倍的軌跡

  • over the years.

    這是快七倍的

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

    這是機器人以比人類快十倍的速度

  • Nowadays, high school students can build robots,

    操作的軌跡

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

    所以說,我們是有辦法讓機器人

  • But, it's very -- now ...

    用比人類快十倍的速度

  • And now, I have a daughter,

    來執行像手術附加工作這樣的精細工作的

  • named Odessa.

    因為這個設計包含了練習和學習

  • She's eight years old.

    多次重覆做同一件事情

  • And she likes robots, too.

    所以這個設計也教了我們一件事情

  • Maybe it runs in the family.

    就是如果你想做好一件事情

  • (Laughter)

    就只能不斷地練習再練習

  • I wish she could meet my dad.

    這些就是這幾年來

  • And now I get to teach her how things work,

    我從機器人身上學到的四門課

  • and we get to build projects together.

    至於機器人學

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

    這個領域也隨著時間進步了許多

  • Robots are the most human of our machines.

    如今就連高中學生

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

    也能做出像我和我父親嘗試做的那種工業機器人

  • but I think they have something important to teach us.

    現在我有個女兒

  • I invite all of you

    名字叫歐妲莎(Odessa)

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

    她現在八歲

  • the machines that you wish for.

    她也喜歡機器人

  • And think about what they might be telling you.

    這大概是家族風氣(笑聲)

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

    我希望她有機會見到我父親

  • the devices we dream about,

    現在換我教她做事情的方法

  • can inspire us to be better humans.

    我們一起做很多東西

  • Thank you.

    我常常想她能夠從中學到什麼

  • (Applause)

    機器人是所有機器中

Translator: Morton Bast Reviewer: Thu-Huong Ha

譯者: Ya-lun Lin 審譯者: James Hung

Subtitles and vocabulary

Click the word to look it up Click the word to find further inforamtion about it

B1 US TED 機器人 花園 天線 設計 父親

【TED】肯-戈德堡。機器人關於做人的4個教訓(Ken Goldberg:機器人關於做人的4個教訓)。 (【TED】Ken Goldberg: 4 lessons from robots about being human (Ken Goldberg: 4 lessons from robots about being human))

  • 32 6
    Zenn posted on 2021/01/14
Video vocabulary