The incredible potential of flexible, soft robots | Giada Gerboni

譯者: Helen Chang 審譯者: S Sung

So, robots.

機器人，

Robots can be programmed

能夠編寫程式以最少出錯的次數

to do the same task millions of times with minimal error,

讓機器人完成數百萬次相同的任務，

something very difficult for us, right?

但對我們人類來說相當困難，對吧？

And it can be very impressive to watch them at work.

看它們工作，可能令人印象很深刻。

Look at them.

看它們。

I could watch them for hours.

我可以連看幾個小時。

No?

不是嗎？

What is less impressive

不太令人印象深刻的是

is that if you take these robots out of the factories,

如果這些機器人到了工廠外，

where the environments are not perfectly known and measured like here,

和這裡全然已知和測量過的 環境不一樣，

to do even a simple task which doesn't require much precision,

即使做個簡單、無須太精密的任務，

this is what can happen.

可能會這樣。

I mean, opening a door, you don't require much precision.

我是說，開一扇門無須太精密。

(Laughter)

（笑聲）

Or a small error in the measurements,

或者測量中的一個小錯誤，

he misses the valve, and that's it --

使它錯過了閥門，就這樣，完了。

(Laughter)

（笑聲）

with no way of recovering, most of the time.

大多數時候沒有辦法恢復。

So why is that?

為什麼呢？

Well, for many years,

多年來，

robots have been designed to emphasize speed and precision,

機器人的設計強調速度和精密度，

and this translates into a very specific architecture.

轉化成為非常具體的架構。

If you take a robot arm,

如果拿一隻機器手臂，

it's a very well-defined set of rigid links

用的是一套很明確的剛性鏈接

and motors, what we call actuators,

和被稱為執行器的機電，

they move the links about the joints.

移動關節周圍的鏈接。

In this robotic structure,

在這機器人結構裡

you have to perfectly measure your environment,

周圍的環境必須被完美地測量，

so what is around,

周圍有什麼，

and you have to perfectly program every movement

也必須完美地編寫

of the robot joints,

每個機器人關節動作的指令，

because a small error can generate a very large fault,

因為一個小錯誤可能會 導致非常大的錯誤，

so you can damage something or you can get your robot damaged

可能損毀東西；

if something is harder.

或者，如果那東西比較硬， 就會損毀機器人。

So let's talk about them a moment.

讓我們來談一會兒，

And don't think about the brains of these robots

不要考慮這些機器人的腦

or how carefully we program them,

或它的程式寫得多仔細，

but rather look at their bodies.

而是看它們的身體。

There is obviously something wrong with it,

身體顯然有點不對勁，

because what makes a robot precise and strong

因為使機器人精確和強大的因素

also makes them ridiculously dangerous and ineffective in the real world,

也使它們在現實世界中 危險和沒效率，

because their body cannot deform

因為它們的身體不能變形，

or better adjust to the interaction with the real world.

也無法更適應 與真實世界的相互作用。

So think about the opposite approach,

因此反向思考，

being softer than anything else around you.

比周圍的其他事物更柔軟。

Well, maybe you think that you're not really able to do anything if you're soft,

也許你認為柔軟就辦不了事。

probably.

或許吧。

Well, nature teaches us the opposite.

然而，大自然告訴我們的恰恰相反。

For example, at the bottom of the ocean,

例如，在海底，

under thousands of pounds of hydrostatic pressure,

在數千磅的靜水壓力下，

a completely soft animal

完全柔軟的動物能移動，

can move and interact with a much stiffer object than him.

也能與比牠硬的物體相互作用。

He walks by carrying around this coconut shell

牠帶著這個椰子殼走來走去，

thanks to the flexibility of his tentacles,

牠的觸手靈活，

which serve as both his feet and hands.

既是腳，也是手。

And apparently, an octopus can also open a jar.

很顯然，章魚能開罐。

It's pretty impressive, right?

令人印象深刻，對吧？

But clearly, this is not enabled just by the brain of this animal,

但顯然，牠辦得到不僅由於腦，

but also by his body,

也由於身體。

and it's a clear example, maybe the clearest example,

這是個明顯，也許最明顯

of embodied intelligence,

展現智能的例子，

which is a kind of intelligence that all living organisms have.

一種生物具有的智能。

We all have that.

我們都有。

Our body, its shape, material and structure,

我們的身體、形狀、材質和結構，

plays a fundamental role during a physical task,

在做動作時起至關重要的作用；

because we can conform to our environment

因為我們符合環境，

so we can succeed in a large variety of situations

因此能在各種情況下成功，

without much planning or calculations ahead.

無需提前計劃或計算。

So why don't we put some of this embodied intelligence

那為什麼不把這些展現的身體智能

into our robotic machines,

放入機器人，

to release them from relying on excessive work

讓它們擺脫

過度依賴計算和感知的工作呢？

on computation and sensing?

Well, to do that, we can follow the strategy of nature,

為了做到這一點， 我們可以遵循自然的戰略；

because with evolution, she's done a pretty good job

因為演化過程中的機器

in designing machines for environment interaction.

與環境的交互作用設計得非常好。

And it's easy to notice that nature uses soft material frequently

容易注意到大自然常用軟質的材料，

and stiff material sparingly.

很少用堅硬的材料。

And this is what is done in this new field of robotics,

這就是在「軟式機器人」

which is called "soft robotics,"

這個新的機器人技術領域裡做的。

in which the main objective is not to make super-precise machines,

主要的目標不是製造超精密的機器，

because we've already got them,

因為我們已經有了；

but to make robots able to face unexpected situations in the real world,

而是要讓機器人能夠面對 現實世界中的意外情況，

so able to go out there.

能夠走出去。

And what makes a robot soft is first of all its compliant body,

要讓機器人柔軟 先要讓的它的身體柔順，

which is made of materials or structures that can undergo very large deformations,

用可承受大變形的材料或結構構成，

so no more rigid links,

不用剛性的連接。

and secondly, to move them, we use what we call distributed actuation,

其次，用分佈式驅動來移動它們，

so we have to control continuously the shape of this very deformable body,

必須不斷地控制 這種變形身體的形狀，

which has the effect of having a lot of links and joints,

這種變形身體 有很多連接和關節的效果，

but we don't have any stiff structure at all.

但沒有任何僵硬的結構。

So you can imagine that building a soft robot is a very different process

可以想像建造軟式機器人 是個非常不一樣的過程，

than stiff robotics, where you have links, gears, screws

不是用鏈接、齒輪、 螺絲的僵硬機器人，

that you must combine in a very defined way.

必須以一種非常明確的方式結合。

In soft robots, you just build your actuator from scratch

做軟式機器人

大多時候只需從頭開始建造執行器，

most of the time,

but you shape your flexible material

但是將柔性的材料

塑造成會回應特定輸入的形式。

to the form that responds to a certain input.

For example, here, you can just deform a structure

例如在這裡，

如果用剛性的鏈接和關節，

doing a fairly complex shape

if you think about doing the same with rigid links and joints,

結構將會相當複雜；

and here, what you use is just one input,

而（軟式結構）這裡只需一個輸入，

such as air pressure.

例如氣壓。

OK, but let's see some cool examples of soft robots.

讓我們看一些軟式機器人的酷例子。

Here is a little cute guy developed at Harvard University,

這裡有個哈佛大學開發的 可愛的小伙子，

and he walks thanks to waves of pressure applied along its body,

由身體上施加的壓力波而行走；

and thanks to the flexibility, he can also sneak under a low bridge,

並且靈活到可以在低矮的橋下潛行，

keep walking,

一直走，

and then keep walking a little bit different afterwards.

一直走，然後有些不同。

And it's a very preliminary prototype,

這是個極為初步的原型，

but they also built a more robust version with power on board

還有個配有電源板的進階版，

that can actually be sent out in the world and face real-world interactions

能實際在現實世界面對面交流，

like a car passing it over it ...

例如汽車開過它的身旁，

and keep working.

它繼續向前走。

It's cute.

它真可愛。

(Laughter)

（笑聲）

Or a robotic fish, which swims like a real fish does in water

還有機器魚，像真魚一樣游在水中，

simply because it has a soft tail with distributed actuation

只因它有柔軟的尾巴，

using still air pressure.

用靜止空氣壓來分佈式驅動它。

That was from MIT,

是麻省理工學院做的。

and of course, we have a robotic octopus.

當然，我們還有機器章魚。

This was actually one of the first projects

實際上是軟式機器人 這新領域開發的第一批專案之一。

developed in this new field of soft robots.

Here, you see the artificial tentacle,

這裡看得到人造的觸手，

but they actually built an entire machine with several tentacles

實際上他們造了 帶有幾隻觸手的整個機器，

they could just throw in the water,

可以把它扔進水中，

and you see that it can kind of go around and do submarine exploration

它可以四處走動，

in a different way than rigid robots would do.

以不同於硬式機器人的 方式在水裡探索，

But this is very important for delicate environments, such as coral reefs.

這對珊瑚礁等微妙環境非常重要。

Let's go back to the ground.

讓我們回到地面。

Here, you see the view

這裡看得到史丹佛大學的同事

from a growing robot developed by my colleagues in Stanford.

正開發的機器人的圖。

You see the camera fixed on top.

相機固定在頂部。

And this robot is particular,

這機器人很特別，

because using air pressure, it grows from the tip,

因為用氣壓，它向上長，

while the rest of the body stays in firm contact with the environment.

而身體的其餘部分 維持與環境的緊密接觸。

And this is inspired by plants, not animals,

它的靈感來自於植物，不是動物，

which grows via the material in a similar manner

植物以類似的方式生長，

so it can face a pretty large variety of situations.

因此能面對各式各樣的狀況。

But I'm a biomedical engineer,

我是生物醫學工程師，

and perhaps the application I like the most

我最喜歡醫學領域的應用，

is in the medical field,

and it's very difficult to imagine a closer interaction with the human body

難以想像還有更為緊密的 與人體的相互作用，

than actually going inside the body,

除非實際進入人體的內部，

for example, to perform a minimally invasive procedure.

例如，執行微創手術。

And here, robots can be very helpful with the surgeon,

在此機器人能對外科醫師很有幫助，

because they must enter the body

因為醫師們必須使用小孔 和直的器械進入人體，

using small holes and straight instruments,

and these instruments must interact with very delicate structures

這些器械必須在很不確定的環境中

in a very uncertain environment,

與非常微妙的結構相互作用，

and this must be done safely.

且必須安全地進行。

Also bringing the camera inside the body,

將相機帶入身體內部，

so bringing the eyes of the surgeon inside the surgical field

將外科醫師的眼睛帶入手術區域

can be very challenging if you use a rigid stick,

如果用剛性棒可能極具挑戰，

像傳統的內視鏡之類的。

like a classic endoscope.

With my previous research group in Europe,

我與以前在歐洲的研究小組一起

we developed this soft camera robot for surgery,

開發這款手術用的軟式照相機器人，

which is very different from a classic endoscope,

與傳統的內視鏡很不同，它能移動，

which can move thanks to the flexibility of the module

這要歸功於模塊的靈活性，

that can bend in every direction and also elongate.

可以向各個方向彎曲或伸長。

And this was actually used by surgeons to see what they were doing

實際上，外科醫師用這種方法

從不同的角度 觀察其他儀器進行的操作，

with other instruments from different points of view,

without caring that much about what was touched around.

無須分心去在乎觸及了什麼。

And here you see the soft robot in action,

在這裡看得到軟式機器人在行動，

and it just goes inside.

它進入體內，

This is a body simulator, not a real human body.

這是一具模擬的人體， 不是真正的人體。

It goes around.

它四處走動。

You have a light, because usually,

有個燈，

you don't have too many lights inside your body.

因為體內通常沒亮光。

We hope.

最好沒有。

(Laughter)

（笑聲）

But sometimes, a surgical procedure can even be done using a single needle,

但有時候，甚至可以 用單針完成外科手術，

and in Stanford now, we are working on a very flexible needle,

我們現今在史丹佛大學 研究一種非常靈活的針頭，

kind of a very tiny soft robot

它是一種非常小巧的軟式機器人，

which is mechanically designed to use the interaction with the tissues

被設計來與組織相互作用，

and steer around inside a solid organ.

在堅實的器官內轉來轉去，

This makes it possible to reach many different targets, such as tumors,

所以通過單個插入孔

就能到達實體器官深處的腫瘤

deep inside a solid organ

by using one single insertion point.

或許多不同目標。

And you can even steer around the structure that you want to avoid

甚至還可以繞過

避開到達目標前的其他結構。

on the way to the target.

So clearly, this is a pretty exciting time for robotics.

顯然對於機器人來說 這是個非常來勁的時刻。

We have robots that have to deal with soft structures,

有了軟式結構的機器人，

so this poses new and very challenging questions

給機器人業界帶來新的、 極具挑戰性的問題，

for the robotics community,

and indeed, we are just starting to learn how to control,

而我們實際上才剛剛 開始學習如何控制，

how to put sensors on these very flexible structures.

如何將傳感器放在這些 非常靈活的結構上。

But of course, we are not even close to what nature figured out

但是當然我們還離自然界

in millions of years of evolution.

數百萬年演變過程中的 發現還遠得很。

But one thing I know for sure:

但有一點我很肯定：

robots will be softer and safer,

機器人會更柔軟、更安全，

and they will be out there helping people.

它們能幫人的還多著呢。

Thank you.

謝謝。

(Applause)

（掌聲）