sponsored links
TEDGlobal 2013

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

June 11, 2013

Ecologist Lian Pin Koh makes a persuasive case for using drones to protect the world's forests and wildlife. These lightweight autonomous flying vehicles can track animals in their natural habitat, monitor the health of rainforests, even combat crime by detecting poachers via thermal imaging. Added bonus? They're also entirely affordable.

Lian Pin Koh - Drones ecologist
Lian Pin Koh expands conservation efforts by championing the use of low-cost autonomous aerial vehicles. Full bio

sponsored links
Double-click the English subtitles below to play the video.
When we think of Nepal,
00:13
we tend to think of the snow-capped mountains
00:15
of the Himalayas,
00:17
the crystal-clear still waters of its alpine lakes,
00:18
or the huge expanse of its grasslands.
00:22
What some of us may not realize
00:25
is that in the Himalayan foothills,
00:27
where the climate is much warmer
00:29
and the landscape much greener,
00:32
there lives a great diversity of wildlife,
00:34
including the one-horned rhinoceros,
00:36
the Asian elephant
00:39
and the Bengal tiger.
00:40
But unfortunately, these animals
00:43
are under constant threat from poachers
00:44
who hunt and kill them for their body parts.
00:47
To stop the killing of these animals,
00:50
battalions of soldiers and rangers
00:53
are sent to protect Nepal's national parks,
00:55
but that is not an easy task,
00:58
because these soldiers have to patrol
01:00
thousands of hectares of forests on foot
01:03
or elephant backs.
01:06
It is also risky for these soldiers
01:07
when they get into gunfights with poachers,
01:10
and therefore Nepal is always looking
01:12
for new ways to help with protecting the forests
01:14
and wildlife.
01:18
Well recently, Nepal acquired a new tool
01:20
in the fight against wildlife crime,
01:22
and these are drones,
01:25
or more specifically, conservation drones.
01:26
For about a year now, my colleagues and I
01:30
have been building drones for Nepal
01:32
and training the park protection personnel
01:34
on the use of these drones.
01:37
Not only does a drone give you
01:39
a bird's-eye view of the landscape,
01:41
but it also allows you to capture detailed,
01:43
high-resolution images of objects on the ground.
01:46
This, for example, is a pair of rhinoceros
01:50
taking a cooling bath on a hot summer day
01:54
in the lowlands of Nepal.
01:57
Now we believe that drones have
01:58
tremendous potential,
02:01
not only for combating wildlife crime,
02:03
but also for monitoring the health
02:05
of these wildlife populations.
02:07
So what is a drone?
02:10
Well, the kind of drone I'm talking about
02:12
is simply a model aircraft
02:14
fitted with an autopilot system,
02:16
and this autopilot unit contains a tiny computer,
02:19
a GPS, a compass, a barometric altimeter
02:23
and a few other sensors.
02:27
Now a drone like this
02:29
is meant to carry a useful payload,
02:31
such as a video camera
02:33
or a photographic camera.
02:35
It also requires a software that allows the user
02:37
to program a mission,
02:40
to tell the drone where to go.
02:42
Now people I talk to are often surprised
02:44
when they hear that these are the only
02:46
four components that make a conservation drone,
02:49
but they are even more surprised
02:52
when I tell them how affordable
these components are.
02:53
The facts is, a conservation drone
02:57
doesn't cost very much more than
02:59
a good laptop computer
03:01
or a decent pair of binoculars.
03:03
So now that you've built
your own conservation drone,
03:06
you probably want to go fly it,
03:09
but how does one fly a drone?
03:12
Well, actually, you don't,
03:14
because the drone flies itself.
03:15
All you have to do is to program a mission
03:18
to tell the drone where to fly.
03:21
But you simply do that by clicking on
03:23
a few way points on the Google Maps interface
03:25
using the open-source software.
03:29
Those missions could be as simple
03:31
as just a few way points,
03:33
or they could be slightly longer
and more complicated,
03:35
to fly along a river system.
03:38
Sometimes, we fly the drone
in a lawnmower-type pattern
03:41
and take pictures of that area,
03:44
and those pictures can be processed
03:47
to produce a map of that forest.
03:49
Other researchers might want to fly the drone
03:51
along the boundaries of a forest
03:54
to watch out for poachers or people
03:56
who might be trying to enter the forest illegally.
03:58
Now whatever your mission is,
04:01
once you've programmed it,
04:03
you simply upload it to the autopilot system,
04:04
bring your drone to the field,
04:07
and launch it simply by tossing it in the air.
04:09
And often we'll go about this mission
04:12
taking pictures or videos along the way,
04:15
and usually at that point,
04:18
we will go grab ourselves a cup of coffee,
04:20
sit back, and relax for the next few minutes,
04:22
although some of us sit back
and panic for the next few minutes
04:25
worrying that the drone will not return.
04:28
Usually it does, and when it does,
04:30
it even lands automatically.
04:33
So what can we do with a conservation drone?
04:36
Well, when we built our first prototype drone,
04:39
our main objective was to fly it over
04:42
a remote rainforest in North Sumatra, Indonesia,
04:45
to look for the nest of a species of great ape
04:49
known as the orangutan.
04:52
The reason we wanted to do that was because
04:54
we needed to know how many individuals
04:57
of this species are still left in that forest.
04:59
Now the traditional method of surveying
05:02
for orangutans is to walk the forest on foot
05:04
carrying heavy equipment
05:08
and to use a pair of binoculars
to look up in the treetops
05:10
where you might find an orangutan or its nest.
05:13
Now as you can imagine,
05:17
that is a very time-consuming, labor-intensive,
05:19
and costly process,
05:22
so we were hoping that drones
05:23
could significantly reduce the cost of surveying
05:25
for orangutan populations in Indonesia
05:28
and elsewhere in Southeast Asia.
05:31
So we were very excited when we captured
05:33
our first pair of orangutan nests on camera.
05:36
And this is it; this is the first ever picture
05:39
of orangutan nests taken with a drone.
05:42
Since then we have taken pictures
05:45
of dozens of these nests
05:47
from around various parts of Southeast Asia,
05:49
and we're now working with computer scientists
05:52
to develop algorithms that can automatically count
05:54
the number of nests from the thousands
05:58
of photos we've collected so far.
06:00
But nests are not the only objects
06:03
these drones can detect.
06:05
This is a wild orangutan
06:07
happily feeding on top of a palm tree,
06:09
seemingly oblivious to our drone
that was flying overhead,
06:11
not once but several times.
06:15
We've also taken pictures of other animals
06:17
including forest buffalos in Gabon,
06:20
elephants, and even turtle nests.
06:24
But besides taking pictures
of just the animals themselves,
06:27
we also take pictures of the habitats
these animals live in,
06:30
because we want to keep track
06:34
of the health of these habitats.
06:36
Sometimes, we zoom out a little
06:39
and look at other things that might be happening
06:41
in the landscape.
06:43
This is an oil palm plantation in Sumatra.
06:45
Now oil palm is a major driver of deforestation
06:48
in that part of the world,
06:51
so we wanted to use this new drone technology
06:53
to keep track of the spread of these plantations
06:56
in Southeast Asia.
06:59
But drones could also be used to keep track of
07:01
illegal logging activities.
07:03
This is a recently logged forest,
07:06
again in Sumatra.
07:08
You could even still see the processed
07:10
wooden planks left on the ground.
07:12
But perhaps the most exciting part
07:15
about taking pictures from the air is
07:18
we could later stitch these pictures together
07:20
using special software to create a map
07:23
of the entire landscape, and this map
07:26
gives us crucial information
07:28
for monitoring land use change,
07:30
to let us know where and when
plantations might be expanding,
07:33
where forests might be contracting,
07:36
or where fires might be breaking out.
07:39
Aerial images could also be processed
07:42
to produce three-dimensional
07:44
computer models of forests.
07:47
Now these models are not just visually appealing,
07:49
but they are also geometrically accurate,
07:52
which means researchers can now measure
07:54
the distance between trees,
07:57
calculate surface area, the volume of vegetation,
07:59
and so on, all of which are important information
08:02
for monitoring the health of these forests.
08:05
Recently, we've also begun experimenting
08:09
with thermal imaging cameras.
08:11
Now these cameras can detect
08:13
heat-emitting objects from the ground,
08:15
and therefore they are very useful
for detecting poachers or their campfires at night.
08:17
So I've told you quite a lot about
08:25
what conservation drones are,
08:26
how you might operate one of these drones,
08:29
and what a drone could do for you.
08:31
I will now tell you where conservation drones
08:33
are being used around the world.
08:35
We built our first prototype drones in Switzerland.
08:38
We brought a few of these to Indonesia
08:41
for the first few test flights.
08:43
Since then, we've been building drones
08:45
for our collaborators from around the world,
08:47
and these include fellow biologists
08:50
and partners from major conservation organizations.
08:52
Perhaps the best and most rewarding part
08:56
about working with these collaborators
08:58
is the feedback they give us
09:00
on how to improve our drones.
09:02
Building drones for us is
09:04
a constant work in progress.
09:06
We are constantly trying to improve them in terms of
09:08
their range, their ruggedness,
09:12
and the amount of payload they can carry.
09:14
We also work with collaborators
09:17
to discover new ways of using these drones.
09:19
For example, camera traps are a common tool
09:22
used by biologists to take pictures of shy animals
09:25
hiding in the forests,
09:29
but these are motion-activated cameras,
09:31
so they snap a picture every time an animal
09:33
crosses their path.
09:36
But the problem with camera traps
09:38
is that the researcher has to go back to the forest
09:39
every so often to retrieve those images,
09:43
and that takes a lot of time,
09:46
especially if there are dozens
09:48
or hundreds of these cameras placed in the forest.
09:50
Now a drone could be designed to perform the task
09:53
much more efficiently.
09:56
This drone, carrying a special sensor,
09:57
could be flown over the forest
10:00
and remotely download these images
10:02
from wi-fi–enabled cameras.
10:05
Radio collars are another tool
10:09
that's commonly used by biologists.
10:11
Now these collars are put onto animals.
10:14
They transmit a radio signal which allows
10:16
the researcher to track the movements
of these animals across the landscape.
10:19
But the traditional way of tracking animals
10:24
is pretty ridiculous,
10:27
because it requires the researcher to be walking
10:29
on the ground carrying a huge
and cumbersome radio antenna,
10:32
not unlike those old TV antennae we used to have
10:36
on our rooftops. Some of us still do.
10:39
A drone could be used to do the same job
10:42
much more efficiently.
10:44
Why not equip a drone
10:46
with a scanning radio receiver,
10:48
fly that over the forest canopy
10:50
in a certain pattern
10:52
which would allow the user or the operator
10:53
to triangulate the location
10:56
of these radio-collared animals remotely
10:59
without having to step foot in the forest.
11:02
A third and perhaps most exciting way
11:05
of using these drones
11:07
is to fly them to a really remote,
11:09
never-explored-before rainforest
11:12
somewhere hidden in the tropics,
11:14
and parachute down a tiny spy microphone
11:16
that would allow us to eavesdrop on the calls
11:20
of mammals, birds, amphibians,
11:22
the Yeti, the Sasquatch, Bigfoot, whatever.
11:25
That would give us biologists
11:30
a pretty good idea of what animals
11:31
might be living in those forests.
11:33
And finally, I would like to show you
11:36
the latest version of our conservation drone.
11:38
The MAJA drone has a wingspan
11:41
of about two meters.
11:43
It weighs only about two kilograms,
11:45
but it can carry half its weight.
11:47
It is a fully autonomous system.
11:49
During its mission, it can even transmit
11:52
a live video feed back to a ground station laptop,
11:54
which allows the user
11:58
to see what the drone is seeing in real time.
12:00
It carries a variety of sensors,
12:03
and the photo quality of some of these sensors
12:05
can be as high as one to two centimeters per pixel.
12:07
This drone can stay in the air for 40 to 60 minutes,
12:12
which gives it a range of up to 50 kilometers.
12:15
That is quite sufficient for most
12:18
of our conservation applications.
12:20
Now, conservation drones began as
12:23
a crazy idea from two biologists
12:25
who are just deeply passionate
about this technology.
12:28
And we believe, strongly believe,
12:31
that drones can and will be a game changer
12:33
for conservation research and applications.
12:36
We've had our fair share of skeptics and critics
12:39
who thought that we were just
fooling around with toy planes.
12:43
And in a way, they are right.
12:46
I mean, let's be honest,
12:48
drones are the ultimate toys for boys.
12:49
But at the same time, we've also gotten to know
12:53
many wonderful colleagues and collaborators
12:55
who share our vision
12:57
and see the potential of conservation drones.
12:59
To us, it is obvious that conservation biologists
13:02
and practitioners should make full use
13:05
of every available tool, including drones,
13:08
in our fight to save the last remaining forests
13:11
and wildlife of this planet.
13:14
Thank you.
13:16
(Applause)
13:18

sponsored links

Lian Pin Koh - Drones ecologist
Lian Pin Koh expands conservation efforts by championing the use of low-cost autonomous aerial vehicles.

Why you should listen

Lian Pin Koh is a relentless tinkerer and science fiction movie geek, though most know him as an environmental scientist. His dreams of combining these interests led him to cofound, with colleague Serge Wich, the site ConservationDrones.org, a project dedicated to gathering intelligence on forests and wildlife through the use of low-cost unmanned flying machines.

Ground surveys are expensive, and are not conducted at a sufficient frequency. Furthermore, some remote tropical forests have never been really surveyed for biodiversity. Koh's machines have already collected valuable information in Sumatra, Congo, Gabon, and Madagascar.

He is an assistant professor of applied ecology and conservation at the ETH Zurich.

The original video is available on TED.com
sponsored links

If you need translations, you can install "Google Translate" extension into your Chrome Browser.
Furthermore, you can change playback rate by installing "Video Speed Controller" extension.

Data provided by TED.

This website is owned and operated by Tokyo English Network.
The developer's blog is here.