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

這是機械蠑螈，

called Pleurodeles waltl.

機械蠑螈是一種遙控機具， 我們設計時，高度模擬了一類

Pleurobot can walk, as you can see here,

稱為歐非肋突螈的兩棲類。

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

你可以看到，機械蠑螈能夠走路，

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

稍後你也會看到，它會游泳，

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

你可能會問，我們為甚麼要 設計這一類的遙控機械?

Indeed, we designed it together with neurobiologists

事實上，這具機械是設計來 作為腦神經研究的科研工具。

to understand how animals move,

我們也確實在設計的時候， 夥同腦神經學家們

and especially how the spinal cord controls locomotion.

去理解動物怎樣律動，

But the more I work in biorobotics,

尤其去理解脊髓如何控制肢體行動。

the more I'm really impressed by animal locomotion.

但當我在機械生物做出愈多的研究，

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

我愈訝異於動物的肢體行動。

or even us as humans,

請你想像一下，一隻海豚的游泳 或一隻貓兒的奔跑跳躍，

when you go jogging or play tennis,

或甚至是我們人類

we do amazing things.

在慢跑或打網球時，

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

我們是在做驚奇的事情。

It has to coordinate more or less 200 muscles perfectly,

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。

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

它需要完美得協調 大約二百組肌肉，

And my goal is to understand how this works.

如果協調得不好， 我們會摔倒，或行動得蹩扭。

There are four main components behind animal locomotion.

而我的目標就是要理解這是如何運作的。

The first component is just the body,

動物的肢體行動， 有四個主要元件。

and in fact we should never underestimate

第一個元件就是身體，

to what extent the biomechanics already simplify locomotion in animals.

事實上，我們都不應該低估

Then you have the spinal cord,

生物力學已經簡化動物的 肢體行動到哪種程度。

and in the spinal cord you find reflexes,

第二個元件就是脊髓，

multiple reflexes that create a sensorimotor coordination loop

你在脊髓裡可以找到反射作用，

between neural activity in the spinal cord and mechanical activity.

多重反射作用會在脊髓的神經活動 與機械活動之間

A third component are central pattern generators.

產生出一種 感知運動的協調迴路，

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

第三個元件是中樞模式產生器。

that can generate, by themselves,

這些非常有趣的迴路， 存在於脊椎動物的脊髓裏，

very coordinated rhythmic patterns of activity

當它接收到非常簡單的輸入訊號時，

while receiving only very simple input signals.

它們可以自己產生出

And these input signals

非常協調且有節奏感的運動，

coming from descending modulation from higher parts of the brain,

而這些輸入訊號，

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

就是從大腦內較高部位 發射出來的「下行性調控」訊號，

will all modulate activity of the spinal cord

就如同運動皮質層、 小腦，基底核一樣，

while we do locomotion.

而當我們在做肢體活動的時候，

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

它能夠調控脊髓的所有活動。

the spinal cord, together with the body,

有趣的是，在某些程度上，

already solve a big part of the locomotion problem.

這些脊髓，連同身體，

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

已經可以解決大部份肢體的活動問題。

it can still run for a while,

你大概也知道一個事實， 當一隻雞被砍頭之後，

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

它還可以再跑一會兒，

already solve a big part of locomotion.

這表示，僅僅較低部位的脊髓和身體，

Now, understanding how this works is very complex,

已經解決了大部份肢體活動的問題。

because first of all,

要理解這是如何運作的， 其實也蠻複雜的，

recording activity in the spinal cord is very difficult.

因為，首先，

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

要記錄脊髓裡面的活動非常困難。

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

在大腦運動皮層植入電極遠比 在脊髓植入電極容易，

Especially in humans, very hard to do.

因為它被脊椎骨保護著。

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

尤其是在人類身上，非常難辦到。

and very dynamic interaction between these four components.

第二個困難，有很大的原因是， 肢體行動在這四個元件之間，

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

是非常複雜且動態交互作用著的。

This is where biorobots like Pleurobot and mathematical models

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。

can really help.

這也是為什麼機械生物，

So what's biorobotics?

在建立像是機械蠑螈和 數學模組上很有幫助的原因。

Biorobotics is a very active field of research in robotics

所以甚麼是機械生物呢？

where people want to take inspiration from animals

機械生物是機械科研裡 一個非常活躍的領域，

to make robots to go outdoors,

人們都想從動物裡得到啟發，

like service robots or search and rescue robots

製成一些可以到户外去的機械人，

or field 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

And the big goal here is to take inspiration from animals

或是農耕機械人。

to make robots that can handle complex terrain --

而主要目的就是， 要從動物身上得到啟發

stairs, mountains, forests,

來製造一些機械人， 可以處理一些複雜的地形--

places where robots still have difficulties

像是樓梯、山脈、森林、

and where animals can do a much better job.

一些機械人仍然遇到困難的地方，

The robot can be a wonderful scientific tool as well.

以及動物可以做得更好的地方。

There are some very nice projects where robots are used,

機械人同樣也是神奇的科研工具，

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

有些很棒的科研項目利用機械人

And this is exactly the purpose of Pleurobot.

做為腦神經、生物力學 或水力學的科研工具。

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

而這就是做機械蠑螈的目的。

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

在我的實驗室，我們夥同腦神經生物學家

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

例如法國波爾多的腦神經生物學家 Jean-Marie Cabelguen，

And here we want to start simple.

我們打算製作出脊髓的模型， 然後在機器人上驗証。

So it's good to start with simple animals

我們希望從簡單出發。

like lampreys, which are very primitive fish,

所以從簡單的動物開始就好，

and then gradually go toward more complex locomotion,

像是七鰓鰻，非常原始的魚類，

like in salamanders,

然後逐漸地邁向更複雜的肢體活動，

but also in cats and in humans,

像是蜥蜴，

in mammals.

但也包含貓、人類，

And here, a robot becomes an interesting tool

哺乳動物等。

to validate our models.

所以，機械人成為了一個

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

可以驗証我們模型的有趣工具。

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

事實上對我來說，機械蠑螈 算是圓了我一個夢想。

making simulations of lamprey and salamander locomotion

大概二十年前，在我博士班的期間，

during my PhD.

我已經在電腦上，製作一些

But I always knew that my simulations were just approximations.

七鰓鰻和蜥蜴肢體活動的模擬，

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

但我一直以來都知道， 我的模擬只是粗略概算。

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

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，

Why not have a real robot and real physics?

這些都是很難在電腦上適當地模擬出來的。

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

為什麼不乾脆做一個 真實的機械人或真實的物體？

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

在眾多的動物裡，蜥蜴是我喜歡的其中之一。

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

你大概想知道為什麼， 因為以兩棲動物而言，

It makes a wonderful link between swimming,

從進化的角度來看， 蜥蜴其實是很重要的動物。

as you find it in eels or fish,

它完美的串聯起

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

水棲動物的游泳 （像是鰻魚或魚）

And in fact, the modern salamander

以及哺乳類動物的四肢活動 （像是貓或人）。

is very close to the first terrestrial vertebrate,

事實上，現代的蜥蜴

so it's almost a living fossil,

與第一代的陸棲脊椎動物非常相近，

which gives us access to our ancestor,

幾乎就是一種活化石，

the ancestor to all terrestrial tetrapods.

讓我們可以接近自己的祖宗，

So the salamander swims

所有陸棲四肢動物的祖宗。

by doing what's called an anguilliform swimming gait,

蜥蜴是藉由一種稱為鰻游的泳態，

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

來進行游泳的動作，

And if you place the salamander on the ground,

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。

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

而當你把蜥蜴放在地面上時，

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

它又會轉化為快走的步態。

which are very nicely coordinated

在這個案例，你有很好的 週期性肢體律動

with this standing wave undulation of the body,

可以非常好地協調出

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

這樣持續性波浪的身體起伏，

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

就如你們現在所看到的 機械蠑螈的步態。

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

事實上，其中一件很令人訝異 卻又讚嘆的事實就是...

So if you take a decerebrated salamander --

這些活動可以僅藉由 脊髓和身體就可以啟動了。

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

所以即使是一隻沒有腦袋的蜥蜴 --

and if you electrically stimulate the spinal cord,

那不是太好, 但當移除了頭顱--

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

而你用電殛刺激脊髓，

If you stimulate a bit more, the gait accelerates.

在低電流的刺激下， 會做出走路一樣的步態。

And at some point, there's a threshold,

如果你稍稍加強刺激度， 步伐就會隨之加快。

and automatically, the animal switches to swimming.

到了若干程度，會有一個臨界點，

This is amazing.

隨後，動物會自動地 從行走轉為游泳

Just changing the global drive,

這真是神乎其技。

as if you are pressing the gas pedal

只是改變了中央的驅動器，

of descending modulation to your spinal cord,

就如同你在踩油門一樣，

makes a complete switch between two very different gaits.

把下行性調控訊號傳遞到你的脊髓，

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

在兩種不一樣的模式間相互切換。

If you stimulate the spinal cord of a cat,

其實同樣的情況， 在貓身上也觀察得到，

you can switch between walk, trot and gallop.

如果你刺激一隻貓的脊髓，

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

你能夠在其間切換模式： 行走、緩跑和急步跑。

at a low level of stimulation,

或在鳥類身上，你可以隨興 切換一隻小鳥，

and flapping its wings at high-level stimulation.

在低電流時，走路，

And this really shows that the spinal cord

在高電流刺激時，揮動翅膀。

is a very sophisticated locomotion controller.

而這告訴我們，

So we studied salamander locomotion in more detail,

脊髓是個非常複雜精密的 肢體行動控制器。

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

於是我們更仔細的研究蜥蜴的肢體行動，

from Professor Martin Fischer in Jena University in Germany.

其我們有一部很好的X光錄影機，

And thanks to that, you really have an amazing machine

是由德國 Jena 大學的 Martin Fischer 教授所提供。

to record all the bone motion in great detail.

感謝有這部神奇的機器，

That's what we did.

把所有的骨骼行動都仔細的紀錄下來。

So we basically figured out which bones are important for us

這就是我們在做的事。

and collected their motion in 3D.

基本上，我們找出了 對我們來說重要的骨骼，

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

並且收集它們的3D動作。

both on ground and in water,

我們所做的就是收集 整個骨骼的動作資料庫，

to really collect a whole database of motor behaviors

從水上到陸上，

that a real animal can do.

實際地去收集一隻動物所有的

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

移動行為資料庫。

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

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。

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

所以我們做了全方位的優化程序 來找出正確的結構、

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

在哪裡放置馬達、

And this is how Pleurobot came to life.

如何把它們連接一起， 盡可能地重製出這些動作等等。

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

機械蠑螈就是這樣成型的。

So what you see here is almost a direct comparison

讓我們來看看它跟 真正的動物有多近似。

between the walking of the real animal and the Pleurobot.

你現在看到的是， 真正的動物和機械蠑螈在行走時

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

直接對比的影片。

of the walking gait.

你可以看到幾乎是一比一的比例，

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

重演着走路的步態。

But even better, we can do swimming.

如果你倒退或慢動作，你可以看得更清楚。

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

更棒的是，我們可以游泳。

(Laughter)

我們甚至為機械蠑螈穿上了潛水衣--

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

（笑聲）

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

然後我們可以到水裡， 開始重製游泳的泳態。

The physics of interaction are complex.

我們對於此很高興， 因為這個真的很難。

Our robot is much bigger than a small animal,

互動的物理現象相當複雜。

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

我們的機械蠑螈要比小動物大很多，

to make sure we had the same interaction physics.

所以我們得找出 稱之為「等比例動態」的頻率，

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

來確定我們也得到了 一樣的互動物理現象。

and we were very, very happy with this.

你可以看到， 最後我們可以非常接近地運動，

So let's go to the spinal cord.

所以我們對此非常非常的高興。

So here what we did with Jean-Marie Cabelguen

現在我們來看看脊髓。

is model the spinal cord circuits.

我們跟 Jean-Marie Cabelguen 一起

And what's interesting is that the salamander

模擬了脊髓的迴路。

has kept a very primitive circuit,

有趣的是，蜥蜴

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

保持了最原始的迴路，

this primitive eel-like fish,

非常相近於我們找到的七鰓鰻，

and it looks like during evolution,

這個像鰻魚的原始魚類，

new neural oscillators have been added to control the limbs,

看起來像是在進化期間，

to do the leg locomotion.

有新的神經振動器 會被加進來去控制肢體

And we know where these neural oscillators are

來帶動腿的行動。

but what we did was to make a mathematical model

我們知道這些神經振動器在哪裡，

to see how they should be coupled

但我們要做的是，計算出數學模式，

to allow this transition between the two very different gaits.

看看怎樣把他們配對起來，

And we tested that on board of a robot.

來讓這兩種非常不同的 動作可以自由轉換。

And this is how it looks.

我們就在機械蠑螈的電板上測試。

So what you see here is a previous version of Pleurobot

而它看起來就像是這樣。

that's completely controlled by our spinal cord model

這裡你們看到的是， 上一代版本的機械蠑螈，

programmed on board of the robot.

完全由我們輸入在電路板上

And the only thing we do

的脊髓模組程式所控制。

is send to the robot through a remote control

我們唯一做的是，

the two descending signals it normally should receive

透過遥控器，向機械人發出

from the upper part of the brain.

兩組下行性調控訊號，而這通常源自於

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

腦部的上半部分。

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

有趣的是，通過這些訊號

For instance,

我們可以完全控制速度、前進、步、泳態。

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

比方說，

and at some point, if we stimulate a lot,

當我們透過低電流作出刺激時， 我們得到的是行走的狀態，

very rapidly it switches to the swimming gait.

來到某種程度， 如果我們加強了刺激 ，

And finally, we can also do turning very nicely

它會迅速地轉化為游泳的狀態。

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

最後，我們也可輕鬆的轉向

And I think it's really beautiful

主要在脊髓左右兩邊， 在其中的一邊加以刺激就可以了。

how nature has distributed control

我覺得這真是漂亮

to really give a lot of responsibility to the spinal cord

自然界先天的分配了控制權

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

把很多責任交付予脊髓，

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

所以大腦的上半部分 不需要再煩惱每一條肌肉。

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

大腦只負擔高層次的調節，

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

協調各肌肉的任務， 就交付予脊髓了。

So this is another project

現在我們來看看貓的行動 和生物力學的重要性。

where we studied cat biomechanics,

這是另一個項目，

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

我們研究貓的生物力學，

And we found three important criteria in the properties,

而我們想知道形態學 對於肢體活動的幫助。

basically, of the limbs.

我們得出了三個性質的標準，

The first one is that a cat limb

基本上，就是肢體內的性質。

more or less looks like a pantograph-like structure.

首先就是貓的肢體，

So a pantograph is a mechanical structure

大概類似導電弓架的結構。

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

導電弓架是一個機電的結構

So a simple geometrical system that kind of coordinates a bit

永恆的保持着 上部份和下部份的平行。

the internal movement of the segments.

其實就是一個簡單的幾何系統，

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

協調着各部位的內部移動。

Most of the muscles are in the trunk,

貓兒肢體的第二個性質是非常輕量。

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

大部份的肌肉集中在驅體內，

and can be moved very rapidly.

這是很棒的點子，因為這樣 肢體不會有低度的惰性

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

反而能夠迅速的活動。

so to handle impacts and forces.

最後很重要的性質是，貓的肢體彈力很強，

And this is how we designed Cheetah-Cub.

有利於處理好衝擊力和震盪力。

So let's invite Cheetah-Cub onstage.

我們也是如此設計小獵豹的。

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

現在有請小獵豹到台上來。

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