A robot that runs and swims like a salamander | Auke Ijspeert

譯者: lisa li 審譯者: 易帆 余

This is Pleurobot.

這是機械蠑螈，

Pleurobot is a robot that we designed to closely mimic a salamander species

機械蠑螈是一種遙控機具， 我們設計時，高度模擬了一類

called Pleurodeles waltl.

稱為歐非肋突螈的兩棲類。

你可以看到，機械蠑螈能夠走路，

Pleurobot can walk, as you can see here,

and as you'll see later, it can also swim.

稍後你也會看到，它會游泳，

你可能會問，我們為甚麼要 設計這一類的遙控機械?

So you might ask, why did we design this robot?

And in fact, this robot has been designed as a scientific tool for neuroscience.

事實上，這具機械是設計來 作為腦神經研究的科研工具。

我們也確實在設計的時候， 夥同腦神經學家們

Indeed, we designed it together with neurobiologists

to understand how animals move,

去理解動物怎樣律動，

and especially how the spinal cord controls locomotion.

尤其去理解脊髓如何控制肢體行動。

但當我在機械生物做出愈多的研究，

But the more I work in biorobotics,

the more I'm really impressed by animal locomotion.

我愈訝異於動物的肢體行動。

If you think of a dolphin swimming or a cat running or jumping around,

請你想像一下，一隻海豚的游泳 或一隻貓兒的奔跑跳躍，

或甚至是我們人類

or even us as humans,

when you go jogging or play tennis,

在慢跑或打網球時，

we do amazing things.

我們是在做驚奇的事情。

And in fact, our nervous system solves a very, very complex control problem.

事實上，我們的神經系統解決了 一個非常非常複雜的控制問題。

It has to coordinate more or less 200 muscles perfectly,

它需要完美得協調 大約二百組肌肉，

如果協調得不好， 我們會摔倒，或行動得蹩扭。

because if the coordination is bad, we fall over or we do bad locomotion.

而我的目標就是要理解這是如何運作的。

And my goal is to understand how this works.

動物的肢體行動， 有四個主要元件。

There are four main components behind animal locomotion.

The first component is just the body,

第一個元件就是身體，

and in fact we should never underestimate

事實上，我們都不應該低估

to what extent the biomechanics already simplify locomotion in animals.

生物力學已經簡化動物的 肢體行動到哪種程度。

第二個元件就是脊髓，

Then you have the spinal cord,

and in the spinal cord you find reflexes,

你在脊髓裡可以找到反射作用，

multiple reflexes that create a sensorimotor coordination loop

多重反射作用會在脊髓的神經活動 與機械活動之間

between neural activity in the spinal cord and mechanical activity.

產生出一種 感知運動的協調迴路，

第三個元件是中樞模式產生器。

A third component are central pattern generators.

These are very interesting circuits in the spinal cord of vertebrate animals

這些非常有趣的迴路， 存在於脊椎動物的脊髓裏，

that can generate, by themselves,

當它接收到非常簡單的輸入訊號時，

very coordinated rhythmic patterns of activity

它們可以自己產生出

while receiving only very simple input signals.

非常協調且有節奏感的運動，

And these input signals

而這些輸入訊號，

coming from descending modulation from higher parts of the brain,

就是從大腦內較高部位 發射出來的「下行性調控」訊號，

like the motor cortex, the cerebellum, the basal ganglia,

就如同運動皮質層、 小腦，基底核一樣，

will all modulate activity of the spinal cord

而當我們在做肢體活動的時候，

while we do locomotion.

它能夠調控脊髓的所有活動。

But what's interesting is to what extent just a low-level component,

有趣的是，在某些程度上，

the spinal cord, together with the body,

這些脊髓，連同身體，

already solve a big part of the locomotion problem.

已經可以解決大部份肢體的活動問題。

You probably know it by the fact that you can cut the head off a chicken,

你大概也知道一個事實， 當一隻雞被砍頭之後，

it can still run for a while,

它還可以再跑一會兒，

showing that just the lower part, spinal cord and body,

這表示，僅僅較低部位的脊髓和身體，

already solve a big part of locomotion.

已經解決了大部份肢體活動的問題。

Now, understanding how this works is very complex,

要理解這是如何運作的， 其實也蠻複雜的，

because first of all,

因為，首先，

recording activity in the spinal cord is very difficult.

要記錄脊髓裡面的活動非常困難。

It's much easier to implant electrodes in the motor cortex

在大腦運動皮層植入電極遠比 在脊髓植入電極容易，

than in the spinal cord, because it's protected by the vertebrae.

因為它被脊椎骨保護著。

Especially in humans, very hard to do.

尤其是在人類身上，非常難辦到。

A second difficulty is that locomotion is really due to a very complex

第二個困難，有很大的原因是， 肢體行動在這四個元件之間，

是非常複雜且動態交互作用著的。

and very dynamic interaction between these four components.

So it's very hard to find out what's the role of each over time.

所以每次要找出那一個元件 擔任那一個角色，真的是很困難。

This is where biorobots like Pleurobot and mathematical models

這也是為什麼機械生物，

can really help.

在建立像是機械蠑螈和 數學模組上很有幫助的原因。

所以甚麼是機械生物呢？

So what's biorobotics?

Biorobotics is a very active field of research in robotics

機械生物是機械科研裡 一個非常活躍的領域，

where people want to take inspiration from animals

人們都想從動物裡得到啟發，

to make robots to go outdoors,

製成一些可以到户外去的機械人，

like service robots or search and rescue robots

像是一些服務業機械人， 或是可從事搜索和救援的機械人

or field robots.

或是農耕機械人。

And the big goal here is to take inspiration from animals

而主要目的就是， 要從動物身上得到啟發

to make robots that can handle complex terrain --

來製造一些機械人， 可以處理一些複雜的地形--

stairs, mountains, forests,

像是樓梯、山脈、森林、

places where robots still have difficulties

一些機械人仍然遇到困難的地方，

and where animals can do a much better job.

以及動物可以做得更好的地方。

The robot can be a wonderful scientific tool as well.

機械人同樣也是神奇的科研工具，

There are some very nice projects where robots are used,

有些很棒的科研項目利用機械人

like a scientific tool for neuroscience, for biomechanics or for hydrodynamics.

做為腦神經、生物力學 或水力學的科研工具。

And this is exactly the purpose of Pleurobot.

而這就是做機械蠑螈的目的。

So what we do in my lab is to collaborate with neurobiologists

在我的實驗室，我們夥同腦神經生物學家

like Jean-Marie Cabelguen, a neurobiologist in Bordeaux in France,

例如法國波爾多的腦神經生物學家 Jean-Marie Cabelguen，

and we want to make spinal cord models and validate them on robots.

我們打算製作出脊髓的模型， 然後在機器人上驗証。

我們希望從簡單出發。

And here we want to start simple.

So it's good to start with simple animals

所以從簡單的動物開始就好，

like lampreys, which are very primitive fish,

像是七鰓鰻，非常原始的魚類，

and then gradually go toward more complex locomotion,

然後逐漸地邁向更複雜的肢體活動，

like in salamanders,

像是蜥蜴，

but also in cats and in humans,

但也包含貓、人類，

in mammals.

哺乳動物等。

And here, a robot becomes an interesting tool

所以，機械人成為了一個

to validate our models.

可以驗証我們模型的有趣工具。

And in fact, for me, Pleurobot is a kind of dream becoming true.

事實上對我來說，機械蠑螈 算是圓了我一個夢想。

Like, more or less 20 years ago I was already working on a computer

大概二十年前，在我博士班的期間，

making simulations of lamprey and salamander locomotion

我已經在電腦上，製作一些

during my PhD.

七鰓鰻和蜥蜴肢體活動的模擬，

But I always knew that my simulations were just approximations.

但我一直以來都知道， 我的模擬只是粗略概算。

Like, simulating the physics in water or with mud or with complex ground,

像是在水中模擬物理現象， 或是在混雜泥土裡或是複雜的地表面上，

it's very hard to simulate that properly on a computer.

這些都是很難在電腦上適當地模擬出來的。

Why not have a real robot and real physics?

為什麼不乾脆做一個 真實的機械人或真實的物體？

So among all these animals, one of my favorites is the salamander.

在眾多的動物裡，蜥蜴是我喜歡的其中之一。

You might ask why, and it's because as an amphibian,

你大概想知道為什麼， 因為以兩棲動物而言，

it's a really key animal from an evolutionary point of view.

從進化的角度來看， 蜥蜴其實是很重要的動物。

It makes a wonderful link between swimming,

它完美的串聯起

as you find it in eels or fish,

水棲動物的游泳 （像是鰻魚或魚）

and quadruped locomotion, as you see in mammals, in cats and humans.

以及哺乳類動物的四肢活動 （像是貓或人）。

事實上，現代的蜥蜴

And in fact, the modern salamander

is very close to the first terrestrial vertebrate,

與第一代的陸棲脊椎動物非常相近，

so it's almost a living fossil,

幾乎就是一種活化石，

which gives us access to our ancestor,

讓我們可以接近自己的祖宗，

the ancestor to all terrestrial tetrapods.

所有陸棲四肢動物的祖宗。

蜥蜴是藉由一種稱為鰻游的泳態，

So the salamander swims

by doing what's called an anguilliform swimming gait,

來進行游泳的動作，

so they propagate a nice traveling wave of muscle activity from head to tail.

它們從頭部到尾部的肌肉活動， 傳遞出一種很優美的游行波浪。

And if you place the salamander on the ground,

而當你把蜥蜴放在地面上時，

it switches to what's called a walking trot gait.

它又會轉化為快走的步態。

In this case, you have nice periodic activation of the limbs

在這個案例，你有很好的 週期性肢體律動

which are very nicely coordinated

可以非常好地協調出

with this standing wave undulation of the body,

這樣持續性波浪的身體起伏，

and that's exactly the gait that you are seeing here on Pleurobot.

就如你們現在所看到的 機械蠑螈的步態。

Now, one thing which is very surprising and fascinating in fact

事實上，其中一件很令人訝異 卻又讚嘆的事實就是...

is the fact that all this can be generated just by the spinal cord and the body.

這些活動可以僅藉由 脊髓和身體就可以啟動了。

So if you take a decerebrated salamander --

所以即使是一隻沒有腦袋的蜥蜴 --

it's not so nice but you remove the head --

那不是太好, 但當移除了頭顱--

and if you electrically stimulate the spinal cord,

而你用電殛刺激脊髓，

at low level of stimulation this will induce a walking-like gait.

在低電流的刺激下， 會做出走路一樣的步態。

If you stimulate a bit more, the gait accelerates.

如果你稍稍加強刺激度， 步伐就會隨之加快。

And at some point, there's a threshold,

到了若干程度，會有一個臨界點，

and automatically, the animal switches to swimming.

隨後，動物會自動地 從行走轉為游泳

This is amazing.

這真是神乎其技。

Just changing the global drive,

只是改變了中央的驅動器，

as if you are pressing the gas pedal

就如同你在踩油門一樣，

of descending modulation to your spinal cord,

把下行性調控訊號傳遞到你的脊髓，

makes a complete switch between two very different gaits.

在兩種不一樣的模式間相互切換。

And in fact, the same has been observed in cats.

其實同樣的情況， 在貓身上也觀察得到，

If you stimulate the spinal cord of a cat,

如果你刺激一隻貓的脊髓，

you can switch between walk, trot and gallop.

你能夠在其間切換模式： 行走、緩跑和急步跑。

Or in birds, you can make a bird switch between walking,

或在鳥類身上，你可以隨興 切換一隻小鳥，

at a low level of stimulation,

在低電流時，走路，

and flapping its wings at high-level stimulation.

在高電流刺激時，揮動翅膀。

And this really shows that the spinal cord

而這告訴我們，

脊髓是個非常複雜精密的 肢體行動控制器。

is a very sophisticated locomotion controller.

So we studied salamander locomotion in more detail,

於是我們更仔細的研究蜥蜴的肢體行動，

and we had in fact access to a very nice X-ray video machine

其我們有一部很好的X光錄影機，

from Professor Martin Fischer in Jena University in Germany.

是由德國 Jena 大學的 Martin Fischer 教授所提供。

And thanks to that, you really have an amazing machine

感謝有這部神奇的機器，

to record all the bone motion in great detail.

把所有的骨骼行動都仔細的紀錄下來。

That's what we did.

這就是我們在做的事。

So we basically figured out which bones are important for us

基本上，我們找出了 對我們來說重要的骨骼，

and collected their motion in 3D.

並且收集它們的3D動作。

And what we did is collect a whole database of motions,

我們所做的就是收集 整個骨骼的動作資料庫，

both on ground and in water,

從水上到陸上，

to really collect a whole database of motor behaviors

實際地去收集一隻動物所有的

that a real animal can do.

移動行為資料庫。

And then our job as roboticists was to replicate that in our robot.

而我們機械設計學家的工作就是， 將這些資料複製到我們的機械人。

So we did a whole optimization process to find out the right structure,

所以我們做了全方位的優化程序 來找出正確的結構、

where to place the motors, how to connect them together,

在哪裡放置馬達、

to be able to replay these motions as well as possible.

如何把它們連接一起， 盡可能地重製出這些動作等等。

And this is how Pleurobot came to life.

機械蠑螈就是這樣成型的。

So let's look at how close it is to the real animal.

讓我們來看看它跟 真正的動物有多近似。

So what you see here is almost a direct comparison

你現在看到的是， 真正的動物和機械蠑螈在行走時

between the walking of the real animal and the Pleurobot.

直接對比的影片。

You can see that we have almost a one-to-one exact replay

你可以看到幾乎是一比一的比例，

of the walking gait.

重演着走路的步態。

If you go backwards and slowly, you see it even better.

如果你倒退或慢動作，你可以看得更清楚。

But even better, we can do swimming.

更棒的是，我們可以游泳。

So for that we have a dry suit that we put all over the robot --

我們甚至為機械蠑螈穿上了潛水衣--

(Laughter)

（笑聲）

and then we can go in water and start replaying the swimming gaits.

然後我們可以到水裡， 開始重製游泳的泳態。

And here, we were very happy, because this is difficult to do.

我們對於此很高興， 因為這個真的很難。

The physics of interaction are complex.

互動的物理現象相當複雜。

Our robot is much bigger than a small animal,

我們的機械蠑螈要比小動物大很多，

so we had to do what's called dynamic scaling of the frequencies

所以我們得找出 稱之為「等比例動態」的頻率，

to make sure we had the same interaction physics.

來確定我們也得到了 一樣的互動物理現象。

But you see at the end, we have a very close match,

你可以看到， 最後我們可以非常接近地運動，

and we were very, very happy with this.

所以我們對此非常非常的高興。

So let's go to the spinal cord.

現在我們來看看脊髓。

So here what we did with Jean-Marie Cabelguen

我們跟 Jean-Marie Cabelguen 一起

is model the spinal cord circuits.

模擬了脊髓的迴路。

有趣的是，蜥蜴

And what's interesting is that the salamander

has kept a very primitive circuit,

保持了最原始的迴路，

which is very similar to the one we find in the lamprey,

非常相近於我們找到的七鰓鰻，

this primitive eel-like fish,

這個像鰻魚的原始魚類，

and it looks like during evolution,

看起來像是在進化期間，

new neural oscillators have been added to control the limbs,

有新的神經振動器 會被加進來去控制肢體

to do the leg locomotion.

來帶動腿的行動。

And we know where these neural oscillators are

我們知道這些神經振動器在哪裡，

but what we did was to make a mathematical model

但我們要做的是，計算出數學模式，

to see how they should be coupled

看看怎樣把他們配對起來，

to allow this transition between the two very different gaits.

來讓這兩種非常不同的 動作可以自由轉換。

And we tested that on board of a robot.

我們就在機械蠑螈的電板上測試。

And this is how it looks.

而它看起來就像是這樣。

So what you see here is a previous version of Pleurobot

這裡你們看到的是， 上一代版本的機械蠑螈，

that's completely controlled by our spinal cord model

完全由我們輸入在電路板上

programmed on board of the robot.

的脊髓模組程式所控制。

And the only thing we do

我們唯一做的是，

is send to the robot through a remote control

透過遥控器，向機械人發出

the two descending signals it normally should receive

兩組下行性調控訊號，而這通常源自於

from the upper part of the brain.

腦部的上半部分。

And what's interesting is, by playing with these signals,

有趣的是，通過這些訊號

we can completely control speed, heading and type of gait.

我們可以完全控制速度、前進、步、泳態。

For instance,

比方說，

when we stimulate at a low level, we have the walking gait,

當我們透過低電流作出刺激時， 我們得到的是行走的狀態，

and at some point, if we stimulate a lot,

來到某種程度， 如果我們加強了刺激 ，

very rapidly it switches to the swimming gait.

它會迅速地轉化為游泳的狀態。

And finally, we can also do turning very nicely

最後，我們也可輕鬆的轉向

by just stimulating more one side of the spinal cord than the other.

主要在脊髓左右兩邊， 在其中的一邊加以刺激就可以了。

And I think it's really beautiful

我覺得這真是漂亮

how nature has distributed control

自然界先天的分配了控制權

to really give a lot of responsibility to the spinal cord

把很多責任交付予脊髓，

so that the upper part of the brain doesn't need to worry about every muscle.

所以大腦的上半部分 不需要再煩惱每一條肌肉。

It just has to worry about this high-level modulation,

大腦只負擔高層次的調節，

and it's really the job of the spinal cord to coordinate all the muscles.

協調各肌肉的任務， 就交付予脊髓了。

So now let's go to cat locomotion and the importance of biomechanics.

現在我們來看看貓的行動 和生物力學的重要性。

So this is another project

這是另一個項目，

where we studied cat biomechanics,

我們研究貓的生物力學，

and we wanted to see how much the morphology helps locomotion.

而我們想知道形態學 對於肢體活動的幫助。

And we found three important criteria in the properties,

我們得出了三個性質的標準，

basically, of the limbs.

基本上，就是肢體內的性質。

The first one is that a cat limb

首先就是貓的肢體，

more or less looks like a pantograph-like structure.

大概類似導電弓架的結構。

So a pantograph is a mechanical structure

導電弓架是一個機電的結構

which keeps the upper segment and the lower segments always parallel.

永恆的保持着 上部份和下部份的平行。

So a simple geometrical system that kind of coordinates a bit

其實就是一個簡單的幾何系統，

the internal movement of the segments.

協調着各部位的內部移動。

A second property of cat limbs is that they are very lightweight.

貓兒肢體的第二個性質是非常輕量。

Most of the muscles are in the trunk,

大部份的肌肉集中在驅體內，

which is a good idea, because then the limbs have low inertia

這是很棒的點子，因為這樣 肢體不會有低度的惰性

and can be moved very rapidly.

反而能夠迅速的活動。

The last final important property is this very elastic behavior of the cat limb,

最後很重要的性質是，貓的肢體彈力很強，

so to handle impacts and forces.

有利於處理好衝擊力和震盪力。

And this is how we designed Cheetah-Cub.

我們也是如此設計小獵豹的。

So let's invite Cheetah-Cub onstage.

現在有請小獵豹到台上來。

So this is Peter Eckert, who does his PhD on this robot,

這位是 Peter Eckert， 他用這部機械人作他的博士學位研究，

and as you see, it's a cute little robot.

你可以看到，這是一隻可愛的小機械。

It looks a bit like a toy,

它看起來有點像是玩具，

but it was really used as a scientific tool

但其實卻是個科研工具，

to investigate these properties of the legs of the cat.

用來查證貓兒四腿的特質。

So you see, it's very compliant, very lightweight,

你可以看到，它非常柔韌，非常輕量，

and also very elastic,

同時也非常的彈性，

so you can easily press it down and it will not break.

所以你可以很輕鬆的 把它壓下而絲毫不損。

It will just jump, in fact.

實際上，它只會彈跳。

And this very elastic property is also very important.

這個彈力的性質也非常重要。

你同時也可以看到一些其他特質，

And you also see a bit these properties

of these three segments of the leg as pantograph.

就是導電弓架作為腳部的三個部份。

Now, what's interesting is that this quite dynamic gait

有趣的是，這很有動態，

is obtained purely in open loop,

純粹來自開放式的迴路，

meaning no sensors, no complex feedback loops.

意思就是沒有探測器， 也沒有複雜的回饋迴路。

And that's interesting, because it means

那是很有趣的，因為那意味著，

that just the mechanics already stabilized this quite rapid gait,

單靠機械模式, 已能把這快速的步態穩定下來，

and that really good mechanics already basically simplify locomotion.

所以只需要有很好的機械性， 基本上已能夠簡化肢體的活動。

To the extent that we can even disturb a bit locomotion,

在某程度上, 我們甚至可以對肢體行動， 作出一些干預，

as you will see in the next video,

就正如你將看到的影像一樣，

where we can for instance do some exercise where we have the robot go down a step,

比方說我們可以做一些運動， 讓機械人下樓梯，

and the robot will not fall over,

而且機械人不會跌倒，

which was a surprise for us.

那其實讓我們很訝異。

This is a small perturbation.

這是輕微的干擾。

I was expecting the robot to immediately fall over,

我預期機械會立刻翻倒，

because there are no sensors, no fast feedback loop.

因為沒有探測器，也沒有迅速回饋的迴路。

But no, just the mechanics stabilized the gait,

但不，單靠機械模式已能把這步態穩定下來，

and the robot doesn't fall over.

而機械人並沒有翻倒。

Obviously, if you make the step bigger, and if you have obstacles,

顯而易見，如果你跨大了步伐， 再加上你遇到了障礙物，

you need the full control loops and reflexes and everything.

你就會需要完整的控制迴路， 還要所有的反射動作及每樣東西。

But what's important here is that just for small perturbation,

這裡重要的是，只是輕微的干擾，

the mechanics are right.

機械性能就正確了。

And I think this is a very important message

我相信這是很重要的訊息，

from biomechanics and robotics to neuroscience,

從生物力學及機械學以至腦神經科學，

saying don't underestimate to what extent the body already helps locomotion.

不要低估身體其實在某種程度， 已經幫助了肢體行動。

Now, how does this relate to human locomotion?

現在，這個跟人類肢體活動有甚麼關係呢？

Clearly, human locomotion is more complex than cat and salamander locomotion,

非常明顯，人類的肢體活動， 遠遠比貓和蜥蜴的複雜，

but at the same time, the nervous system of humans is very similar

但同時間，人類的神經系統

to that of other vertebrates.

卻又和其他的脊椎動物非常類近。

And especially the spinal cord

尤其是脊髓，

is also the key controller for locomotion in humans.

同樣也是人類肢體行動的關鍵控制器。

That's why, if there's a lesion of the spinal cord,

所以，這也是為什麼當脊髓有損害時，

this has dramatic effects.

會帶來很嚴重的後果。

The person can become paraplegic or tetraplegic.

那個人可能會變成 下半身癱瘓或四肢癱瘓。

This is because the brain loses this communication

這是因為大腦失去了

with the spinal cord.

與脊髓的溝通。

Especially, it loses this descending modulation

尤其是大腦失去了啟動

to initiate and modulate locomotion.

和協調肢體活動作用的下行性調控訊號時。

So a big goal of neuroprosthetics

所以神經義肢的一個很大的目標

is to be able to reactivate that communication

就是可以重啟這個溝通，

using electrical or chemical stimulations.

借助電子或化學的刺激。

And there are several teams in the world that do exactly that,

世界上有好幾個隊伍 正在進行這樣的項目，

especially at EPFL.

尤其是 EPFL 的隊伍。

My colleagues Grégoire Courtine and Silvestro Micera,

和我一起合作的同事

with whom I collaborate.

Grégoire Courtine 和 Silvestro Micera 。

But to do this properly, it's very important to understand

要恰當地做好這事情，務必要明白

how the spinal cord works,

脊髓如何運作，

how it interacts with the body,

它怎樣和身體互動，

and how the brain communicates with the spinal cord.

而大腦又怎樣和脊髓溝通。

This is where the robots and models that I've presented today

今天我向各位展示的機械和模型

will hopefully play a key role

希望就是在這些非常重要的目標上

towards these very important goals.

可以做出貢獻。

Thank you.

謝謝。

(Applause)

(掌聲)

Bruno Giussani: Auke, I've seen in your lab other robots

Bruno Giussani: 奧克, 我曾經在 你的實驗室裡，看過其他的機械人，

that do things like swim in pollution

做一些例如在污染環境中游泳的事情，

and measure the pollution while they swim.

並在游泳期間測量污染程度。

But for this one,

但這一個，

you mentioned in your talk, like a side project,

你在演講時表示，這只是一個子項目，

search and rescue,

負責搜索和救援，

and it does have a camera on its nose.

它在鼻子上也安裝了照相機。

Auke Ijspeert: Absolutely. So the robot --

奧克．捷思皮爾：對的。所以這個機械人--

We have some spin-off projects

我們有一些是分拆開來的項目

where we would like to use the robots to do search and rescue inspection,

我們希望主要用作 搜索和救援的調察，

so this robot is now seeing you.

所以這位機械人現正看着你。

And the big dream is to, if you have a difficult situation

而遠大的夢想就是，一旦你遭遇到困難

like a collapsed building or a building that is flooded,

例如倒塌的建築物， 或是被淹沒的建築物，

and this is very dangerous for a rescue team or even rescue dogs,

這樣的情況，對救援隊、甚至救援狗而言 都是非常危險的，

why not send in a robot that can crawl around, swim, walk,

為什麼不用可以四處爬行、游泳 和行走的機械人取替，

with a camera onboard to do inspection and identify survivors

並加上照相機來協助調察和找出生還者，

甚至有可能與生還者建立溝通。

and possibly create a communication link with the survivor.

BG: 當然，假設生還者沒有被它的外型嚇到。

BG: Of course, assuming the survivors don't get scared by the shape of this.

AI: Yeah, we should probably change the appearance quite a bit,

AI: 對，我們大概也需要 把它的外型改良一下，

不然生還者 可能會心臟病發而死亡，

because here I guess a survivor might die of a heart attack

just of being worried that this would feed on you.

僅是擔心它會把你吃掉。

But by changing the appearance and it making it more robust,

把它的外型改良一下， 同時令它更耐用，

I'm sure we can make a good tool out of it.

我深信這能夠成為很好的工具。

BG: Thank you very much. Thank you and your team.

非常感謝。謝謝你和你的團隊。