MyOOPS開放式課程
請加入會員以使用更多個人化功能
來自全球頂尖大學的開放式課程,現在由世界各國的數千名義工志工為您翻譯成中文。請免費享用!
課程來源:TED
     

 

Lian Pin Koh 談保育型無人機的鳥瞰景觀

Lian Pin Koh: A drone's-eye view of conservation

 

Photo of three lions hunting on the Serengeti.

講者:Lian Pin Koh

2013年6月演講,2013年11月在TEDGlobal 2013上線

 

翻譯:洪曉慧

編輯:朱學恆

簡繁轉換:洪曉慧

後製:洪曉慧

字幕影片後制:謝旻均

 

影片請按此下載

MAC及手持裝置版本請按此下載

閱讀中文字幕純文字版本

 

關於這場演講

生態學家Lian Pin Koh藉由極具說服力的實例,闡述使用無人機保護世上森林及野生動物的方法。這些輕巧的自動駕駛飛行器可追蹤生存於自然棲地的動物、監測雨林健康狀態,甚至可藉由熱像儀偵測盜獵者以打擊犯罪。更棒的是?成本十分低廉。

 

關於Lian Pin Koh

Lian Pin Koh藉由提倡使用低成本自主飛行器擴展保育工作的實行。

 

為什麼要聽他演講

Lian Pin Koh是不屈不撓的發明家及科幻片狂熱者,雖然在大多數人眼中他是環境科學家。他將這些興趣結合的夢想,導致他與同事Serge Wich共同成立ConservationDrones.org,這是使用低成本無人飛行器收集森林及野生動物資訊的計畫。

 

地面調查相當昂貴,且無法以足夠的頻率進行。此外,某些偏遠的熱帶森林從未真正進行過生物多樣性調查。Koh的機器已在蘇門答臘、剛果、加蓬及馬達加斯加收集到珍貴的資訊。

 

他是蘇黎世聯邦理工學院應用生態學及保育助理教授。

 

Lian Pin Koh的英語網上資料

Web: Lian Pin Koh

School: ETH Zurich

YouTube: Lian Pin Koh

Twitter: @lianpinkoh

Facebook: lian.pin.koh

 

[TED科技‧娛樂‧設計]

已有中譯字幕的TED影片目錄(繁體)(簡體)。請注意繁簡目錄是不一樣的。

 

Lian Pin Koh 談保育型無人機的鳥瞰景觀

 

當我們想起尼泊爾時,往往會想到冰雪覆蓋的喜馬拉雅山、如水晶般清澈而平靜的高山湖、或綿延不盡的草原。有些人或許不曾意識到,在喜馬拉雅山山麓,氣候更加溫暖、景色更青翠宜人的地方,住著種類繁多的野生動物,包括印度犀牛、亞洲象、孟加拉虎。但不幸的是,這些動物長期面臨盜獵者的威脅,為了獲取牠們身上某部分而進行獵殺。為了遏止這種殺戮動物的行為,大量士兵及巡守員被派遣去保護尼泊爾國家公園。但這並非輕鬆的任務,因為士兵們必須徒步或騎大象巡邏數千公頃的森林。對士兵來說這也是風險極高的任務,當他們與盜獵者交火時。因此尼泊爾一直在尋找保護森林及野生動物的新方法。

 

最近尼泊爾獲得一項打擊盜獵野生動物罪行的新工具,就是無人駕駛飛機;或更確切地說,保育型無人機。大約一年的時間當中,我和同事為尼泊爾製造無人機,並訓練公園保護員使用無人機。無人機不僅能提供景觀鳥瞰圖,也可使你捕捉高解析度地面物體影像。例如這是一對犀牛,在炎炎夏日中,於尼泊爾一處低地泡冷水澡。我們相信無人機擁有巨大潛力,不僅可用來打擊盜獵行為,也可監測野生動物族群的健康狀況。

 

因此,無人機到底是什麼?好,我所說的無人機,只是配備了自動駕駛系統的模型飛機。這個自動駕駛系統包含微型電腦、GPS、指南針、氣壓高度計和一些其他感測器。像這樣的無人機必須具備有效載荷,例如一台攝影機或相機。它也需要軟體,讓使用者編寫任務程式,告訴無人機該飛往何處。

 

人們總是驚訝不已,當他們聽見僅需這四個部分就能組成一架保育型無人機。但更令他們驚訝的是,當我告訴他們這些組件多麼平價時。事實上,保育型無人機的價格不會比高階筆電或高級雙筒望遠鏡貴多少。

 

因此如果你製造了一架保育型無人機,你或許想讓它飛行,但如何操控一架無人機?好,事實上你不需要這麼做,因為無人機可自行飛行。你只需要設計一個任務,告訴無人機該飛往何處。你只需要使用開放源碼軟體,在Google地圖介面上點擊幾個航點。這些任務可簡單到僅包含幾個航點,或者也可以稍長、稍複雜些,例如沿著一條河系飛行。有時我們讓無人機以「割草機」模式飛行,拍攝那個區域的照片。這些照片可進行處理,以製作那片森林的地圖。有些研究者或許想讓無人機沿森林邊緣飛行,監視盜獵者或試圖非法進入森林的人。

 

無論你的目的為何,一旦寫好程式,只需將它上傳至自動駕駛系統,將無人機帶到空曠的地方,將它擲向空中,即可起飛。我們通常會在任務執行過程中拍照或錄影,通常這個時候我們會替自己拿杯咖啡、靠在椅子上,放鬆幾分鐘。儘管有些人坐下後緊張地度過幾分鐘,擔心無人機一去不回。通常它都能順利返航,甚至自動著陸。

 

因此我們能藉由保育型無人機做什麼?好,當我們製造出第一架無人駕駛原型機時,主要目的是讓它飛越印尼北蘇門答臘一處偏遠的雨林,尋找一種名為紅毛猩猩的大型猿類巢穴。我們想這麼做的原因是,我們需要知道那片雨林中還剩下多少隻這種猿類。搜尋紅毛猩猩的傳統方法是徒步穿越雨林,背著沉重的裝備,用雙筒望遠鏡在樹梢尋找紅毛猩猩或牠的巢穴可能存在之處。你可以想像這是多麼費時、費力、費錢的過程,因此我們希望無人機能大幅減少調查印尼及東南亞其他地區紅毛猩猩數量的成本。因此拍攝到第一對紅毛猩猩的巢穴時,我們非常興奮。就是這張,首次藉由無人機拍攝到的紅毛猩猩巢穴照片。從那時起,我們拍攝了很多類似的巢穴相片,遍及東南亞各個地區。現在我們正與電腦科學家合作,開發某種演算法,可藉由我們至今蒐集的數千張相片自動計算巢穴數量。

 

但這些巢穴並非無人機可偵測到的唯一物體。這是一隻野生紅毛猩猩,在棕櫚樹梢開心地進食,似乎不曾察覺我們的無人機正飛越牠的頭頂。不只一次,而是很多次。我們也拍攝了其他動物的照片,包括加彭的非洲水牛、大象、甚至烏龜巢穴。

 

但除了拍攝野生動物照片,我們也拍攝這些動物的棲息地,因為我們想追蹤這些棲息地的生態情況。有時我們稍微將鏡頭拉遠些,觀察其他可能發生在這片區域的事。這是位於蘇門答臘的油棕櫚園,目前該地區的油棕櫚種植是造成森林砍伐的主要元兇,因此我們想利用這種無人機新技術,追蹤這些種植園在東南亞的擴散速度。但無人機也可用於追蹤非法伐木活動。這是最近遭到砍伐的森林,同樣在蘇門答臘,你甚至可以看見地上殘留著加工過的木板。

 

但這些空照圖最令人興奮的部分也許是,我們可用特殊軟體將這些照片組合在一起,製作一張全景地圖。這張地圖可提供關於土地使用狀況改變的關鍵資訊,讓我們知道何時、何處的種植園可能正在擴張,何處的森林可能正在縮小,或何處可能正發生火災。這些空照圖也可經過處理,製作成森林的3D電腦模型。這些模型不僅賞心悅目,也擁有精確的幾何比例,意味著研究員可藉此測量樹木間的距離、計算表面積、植被體積等等,這些都是監測森林生態情況的重要資訊。最近我們也開始使用熱像儀進行測試。這些儀器可偵測地面上的熱源,因此對於偵測盜獵者或他們夜晚所升的營火非常有用。

 

因此我已告訴你們許多關於保育型無人機的資訊:如何操作無人機,以及無人機能為你做什麼。現在我要告訴各位世上哪些地方正使用保育型無人機。我們在瑞士製造了第一批無人駕駛原型機,我們帶了一些前往印尼,進行首次試飛。從那時起,我們為世界各地的合作夥伴製造無人機,包括生物學家及大型保育機構的合作夥伴。

 

與這些夥伴合作的過程中,最棒、最有價值的部分也許在於他們給予我們如何改進無人機的回饋意見。對我們來說,製造無人機是不斷改進的過程。我們致力於改進它們的飛行範圍、耐用性以及可承受的有效載荷,我們也與合作夥伴共同開發運用無人機的新方法。例如相機陷阱是生物學家常用的工具,用來拍攝藏身於樹林中的害羞動物照片。但這些是動作感應式相機,每當動物經過這條途徑就會拍照。但相機陷阱的問題在於,每隔一段時間研究員就得返回森林取照片,這相當耗時,特別是有數十台或數百台相機放置在樹林裡時。現在無人機可設計成以更有效率的方式執行這個任務。這種攜帶特殊感應器的無人機可飛越樹林上空,藉由能發射wi-fi訊號的相機遠程下載這些相片。

 

無線追蹤項圈是另一種生物學家常用的工具。這些項圈可放置在動物身上,它們可傳輸某種無線訊號,使研究者得以追蹤這片區域中的動物行蹤。但這種傳統的追蹤動物方法十分可笑,因為研究者必須攜帶巨大而笨重的無線電接收裝置四處行走,就像以往裝設在屋頂上的老式電視天線,有些人或許仍在使用。無人機可以更有效率的方式進行同樣的工作。為何不在無人機上裝設掃描式無線電接收器,讓它以特定飛行模式飛越樹林上空,讓使用者或操縱者得以藉由三角測量法,為這些戴著無線追蹤項圈的動物進行遠端定位,而不需進入森林。

 

第三種,也許是最令人振奮的無人機使用方法,就是讓它們飛到相當偏遠、隱藏在熱帶地區、不曾被探索過的雨林,然後空投一個微型麥克風,使我們能聽見哺乳類、鳥類及兩棲類動物的聲音,甚至雪人、野人、大腳怪等任何聲音。這將使生物學家得到相當有用的概念,關於那些森林裡可能存在什麼動物。

 

最後,我想向各位展示我們最新型的保育型無人機。這款MAJA無人機翼長大約兩公尺,重量大約只有兩公斤,但它可攜帶相當於本身重量一半的裝備,擁有全自動系統。執行任務過程中,它甚至可將現場影像傳回地面接收站,讓使用者得以即時目睹無人機拍攝的畫面。它攜帶了各式各樣的感應器,某些感應器的相片畫質可高達每像素1至2釐米。這架無人機可連續飛行40至60分鐘,範圍可達50公里,這對大部分保育工作來說已十分足夠。

 

好,保育型無人機起源於兩位生物學家的瘋狂構想,他們十分醉心於這項科技。我們相信、深信不疑:無人機將使保育研究及相關應用大幅改觀。我們遭受過不少質疑與批評,有人認為我們只是拿玩具飛機胡搞。以某種程度而言,他們說的沒錯。我的意思是,坦白說,無人機確實是男孩夢寐以求的玩具。但同時,我們也必須瞭解,許多優秀的同事及合作夥伴擁有和我們相同的願景,並看出保育型無人機的潛力。對我們而言,顯然保育生物學家和環保實踐者都應充分利用每一項可用的工具,包括無人機,盡力拯救這個星球僅存的樹林和野生動物。

 

謝謝。

 

(掌聲)

 

以下為系統擷取之英文原文

About the talk

生態學家高良品為使用無人機來保護世界上的森林及野生動物,做出極具說服力的實例。這些輕巧的自動駕駛飛行器,可以追蹤在自然棲地內的動物,監測雨林的健康狀態,甚至可以經由熱感顯像偵測盜獵者以打擊犯罪。而且,它們經濟實惠。
 
About Lian Pin Koh
Jared Diamond investigates why cultures prosper or decline -- and what we can learn by taking a broad look across many kinds of societies.
 
About the transcript
When we think of Nepal, we tend to think of the snow-capped mountains of the Himalayas, the crystal-clear still waters of its alpine lakes, or the huge expanse of its grasslands. What some of us may not realize is that in the Himalayan foothills, where the climate is much warmer and the landscape much greener, there lives a great diversity of wildlife, including the one-horned rhinoceros, the Asian elephant and the Bengal tiger. But unfortunately, these animals are under constant threat from poachers who hunt and kill them for their body parts. To stop the killing of these animals, battalions of soldiers and rangers are sent to protect Nepal's national parks, but that is not an easy task, because these soldiers have to patrol thousands of hectares of forests on foot or elephant backs. It is also risky for these soldiers when they get into gunfights with poachers, and therefore Nepal is always looking for new ways to help with protecting the forests and wildlife.
 
Well recently, Nepal acquired a new tool in the fight against wildlife crime, and these are drones, or more specifically, conservation drones. For about a year now, my colleagues and I have been building drones for Nepal and training the park protection personnel on the use of these drones. Not only does a drone give you a bird's-eye view of the landscape, but it also allows you to capture detailed, high-resolution images of objects on the ground. This, for example, is a pair of rhinoceros taking a cooling bath on a hot summer day in the lowlands of Nepal. Now we believe that drones have tremendous potential, not only for combating wildlife crime, but also for monitoring the health of these wildlife populations.
 
So what is a drone? Well, the kind of drone I'm talking about is simply a model aircraft fitted with an autopilot system, and this autopilot unit contains a tiny computer, a GPS, a compass, a barometric altimeter and a few other sensors. Now a drone like this is meant to carry a useful payload, such as a video camera or a photographic camera. It also requires a software that allows the user to program a mission, to tell the drone where to go.
 
Now people I talk to are often surprised when they hear that these are the only four components that make a conservation drone, but they are even more surprised when I tell them how affordable these components are. The facts is, a conservation drone doesn't cost very much more than a good laptop computer or a decent pair of binoculars.
 
So now that you've built your own conservation drone, you probably want to go fly it, but how does one fly a drone? Well, actually, you don't, because the drone flies itself. All you have to do is to program a mission to tell the drone where to fly. But you simply do that by clicking on a few way points on the Google Maps interface using the open-source software. Those missions could be as simple as just a few way points, or they could be slightly longer and more complicated, to fly along a river system. Sometimes, we fly the drone in a lawnmower-type pattern and take pictures of that area, and those pictures can be processed to produce a map of that forest. Other researchers might want to fly the drone along the boundaries of a forest to watch out for poachers or people who might be trying to enter the forest illegally.
 
Now whatever your mission is, once you've programmed it, you simply upload it to the autopilot system, bring your drone to the field, and launch it simply by tossing it in the air. And often we'll go about this mission taking pictures or videos along the way, and usually at that point, we will go grab ourselves a cup of coffee, sit back, and relax for the next few minutes, although some of us sit back and panic for the next few minutes worrying that the drone will not return. Usually it does, and when it does, it even lands automatically.
 
So what can we do with a conservation drone? Well, when we built our first prototype drone, our main objective was to fly it over a remote rainforest in North Sumatra, Indonesia, to look for the nest of a species of great ape known as the orangutan. The reason we wanted to do that was because we needed to know how many individuals of this species are still left in that forest. Now the traditional method of surveying for orangutans is to walk the forest on foot carrying heavy equipment and to use a pair of binoculars to look up in the treetops where you might find an orangutan or its nest. Now as you can imagine, that is a very time-consuming, labor-intensive, and costly process, so we were hoping that drones could significantly reduce the cost of surveying for orangutan populations in Indonesia and elsewhere in Southeast Asia. So we were very excited when we captured our first pair of orangutan nests on camera. And this is it; this is the first ever picture of orangutan nests taken with a drone. Since then we have taken pictures of dozens of these nests from around various parts of Southeast Asia, and we're now working with computer scientists to develop algorithms that can automatically count the number of nests from the thousands of photos we've collected so far.
 
But nests are not the only objects these drones can detect. This is a wild orangutan happily feeding on top of a palm tree, seemingly oblivious to our drone that was flying overhead, not once but several times. We've also taken pictures of other animals including forest buffalos in Gabon, elephants, and even turtle nests.
 
But besides taking pictures of just the animals themselves, we also take pictures of the habitats these animals live in, because we want to keep track of the health of these habitats. Sometimes, we zoom out a little and look at other things that might be happening in the landscape. This is an oil palm plantation in Sumatra. Now oil palm is a major driver of deforestation in that part of the world, so we wanted to use this new drone technology to keep track of the spread of these plantations in Southeast Asia. But drones could also be used to keep track of illegal logging activities. This is a recently logged forest, again in Sumatra. You could even still see the processed wooden planks left on the ground.
 
But perhaps the most exciting part about taking pictures from the air is we could later stitch these pictures together using special software to create a map of the entire landscape, and this map gives us crucial information for monitoring land use change, to let us know where and when plantations might be expanding, where forests might be contracting, or where fires might be breaking out. Aerial images could also be processed to produce three-dimensional computer models of forests. Now these models are not just visually appealing, but they are also geometrically accurate, which means researchers can now measure the distance between trees, calculate surface area, the volume of vegetation, and so on, all of which are important information for monitoring the health of these forests. Recently, we've also begun experimenting with thermal imaging cameras. Now these cameras can detect heat-emitting objects from the ground, and therefore they are very useful for detecting poachers or their campfires at night.
 
So I've told you quite a lot about what conservation drones are, how you might operate one of these drones, and what a drone could do for you. I will now tell you where conservation drones are being used around the world. We built our first prototype drones in Switzerland. We brought a few of these to Indonesia for the first few test flights. Since then, we've been building drones for our collaborators from around the world, and these include fellow biologists and partners from major conservation organizations.
 
Perhaps the best and most rewarding part about working with these collaborators is the feedback they give us on how to improve our drones. Building drones for us is a constant work in progress. We are constantly trying to improve them in terms of their range, their ruggedness, and the amount of payload they can carry. We also work with collaborators to discover new ways of using these drones. For example, camera traps are a common tool used by biologists to take pictures of shy animals hiding in the forests, but these are motion-activated cameras, so they snap a picture every time an animal crosses their path. But the problem with camera traps is that the researcher has to go back to the forest every so often to retrieve those images, and that takes a lot of time, especially if there are dozens or hundreds of these cameras placed in the forest. Now a drone could be designed to perform the task much more efficiently. This drone, carrying a special sensor, could be flown over the forest and remotely download these images from wi-fi–enabled cameras.
 
Radio collars are another tool that's commonly used by biologists. Now these collars are put onto animals. They transmit a radio signal which allows the researcher to track the movements of these animals across the landscape. But the traditional way of tracking animals is pretty ridiculous, because it requires the researcher to be walking on the ground carrying a huge and cumbersome radio antenna, not unlike those old TV antennae we used to have on our rooftops. Some of us still do. A drone could be used to do the same job much more efficiently. Why not equip a drone with a scanning radio receiver, fly that over the forest canopy in a certain pattern which would allow the user or the operator to triangulate the location of these radio-collared animals remotely without having to step foot in the forest.
 
A third and perhaps most exciting way of using these drones is to fly them to a really remote, never-explored-before rainforest somewhere hidden in the tropics, and parachute down a tiny spy microphone that would allow us to eavesdrop on the calls of mammals, birds, amphibians, the Yeti, the Sasquatch, Bigfoot, whatever. That would give us biologists a pretty good idea of what animals might be living in those forests.
 
And finally, I would like to show you the latest version of our conservation drone. The MAJA drone has a wingspan of about two meters. It weighs only about two kilograms, but it can carry half its weight. It is a fully autonomous system. During its mission, it can even transmit a live video feed back to a ground station laptop, which allows the user to see what the drone is seeing in real time. It carries a variety of sensors, and the photo quality of some of these sensors can be as high as one to two centimeters per pixel. This drone can stay in the air for 40 to 60 minutes, which gives it a range of up to 50 kilometers. That is quite sufficient for most of our conservation applications.
 
Now, conservation drones began as a crazy idea from two biologists who are just deeply passionate about this technology. And we believe, strongly believe, that drones can and will be a game changer for conservation research and applications. We've had our fair share of skeptics and critics who thought that we were just fooling around with toy planes. And in a way, they are right. I mean, let's be honest, drones are the ultimate toys for boys. But at the same time, we've also gotten to know many wonderful colleagues and collaborators who share our vision and see the potential of conservation drones. To us, it is obvious that conservation biologists and practitioners should make full use of every available tool, including drones, in our fight to save the last remaining forests and wildlife of this planet.
 
Thank you.
 
(Applause)

留下您對本課程的評論
標題:
您目前為非會員,留言名稱將顯示「匿名非會員」
只能進行20字留言

留言內容:

驗證碼請輸入1 + 2 =

標籤

現有標籤:1
新增標籤:


有關本課程的討論

目前暫無評論,快來留言吧!

Creative Commons授權條款 本站一切著作係採用 Creative Commons 授權條款授權。
協助推廣單位: