TED2016
Raffaello D'Andrea: Meet the dazzling flying machines of the future
Filmed:
Readability: 5
8,053,974 views
When you hear the word "drone," you probably think of something either very useful or very scary. But could they have aesthetic value? Autonomous systems expert Raffaello D'Andrea develops flying machines, and his latest projects are pushing the boundaries of autonomous flight -- from a flying wing that can hover and recover from disturbance to an eight-propeller craft that's ambivalent to orientation ... to a swarm of tiny coordinated micro-quadcopters. Prepare to be dazzled by a dreamy, swirling array of flying machines as they dance like fireflies above the TED stage.
Raffaello D'Andrea - Autonomous systems pioneer
Raffaello D'Andrea explores the possibilities of autonomous technology by collaborating with artists, engineers and entrepreneurs. Full bio
Raffaello D'Andrea explores the possibilities of autonomous technology by collaborating with artists, engineers and entrepreneurs. Full bio
Double-click the English transcript below to play the video.
00:12
What started as a platform for hobbyists
0
920
1976
00:14
is poised to become
a multibillion-dollar industry.
a multibillion-dollar industry.
1
2920
2936
00:17
Inspection, environmental monitoring,
photography and film and journalism:
photography and film and journalism:
2
5880
4056
00:21
these are some of the potential
applications for commercial drones,
applications for commercial drones,
3
9960
3176
00:25
and their enablers
are the capabilities being developed
are the capabilities being developed
4
13160
2616
00:27
at research facilities around the world.
5
15800
1936
00:29
For example, before aerial
package delivery
package delivery
6
17760
2616
00:32
entered our social consciousness,
7
20400
1816
00:34
an autonomous fleet of flying machines
built a six-meter-tall tower
built a six-meter-tall tower
8
22240
3976
00:38
composed of 1,500 bricks
9
26240
1936
00:40
in front of a live audience
at the FRAC Centre in France,
at the FRAC Centre in France,
10
28200
2936
00:43
and several years ago,
they started to fly with ropes.
they started to fly with ropes.
11
31160
2576
00:45
By tethering flying machines,
12
33760
1416
00:47
they can achieve high speeds
and accelerations in very tight spaces.
and accelerations in very tight spaces.
13
35200
3776
00:51
They can also autonomously build
tensile structures.
tensile structures.
14
39000
3096
00:54
Skills learned include how to carry loads,
15
42120
2416
00:56
how to cope with disturbances,
16
44560
1456
00:58
and in general, how to interact
with the physical world.
with the physical world.
17
46040
3096
01:01
Today we want to show you some
new projects that we've been working on.
new projects that we've been working on.
18
49160
3336
01:04
Their aim is to push the boundary
of what can be achieved
of what can be achieved
19
52520
2936
01:07
with autonomous flight.
20
55480
1536
01:09
Now, for a system to function
autonomously,
autonomously,
21
57040
2456
01:11
it must collectively know the location
of its mobile objects in space.
of its mobile objects in space.
22
59520
4456
01:16
Back at our lab at ETH Zurich,
23
64000
1896
01:17
we often use external cameras
to locate objects,
to locate objects,
24
65920
2856
01:20
which then allows us to focus our efforts
25
68800
1976
01:22
on the rapid development
of highly dynamic tasks.
of highly dynamic tasks.
26
70800
2656
01:25
For the demos you will see today, however,
27
73480
2056
01:27
we will use new localization technology
developed by Verity Studios,
developed by Verity Studios,
28
75560
3296
01:30
a spin-off from our lab.
29
78880
1616
01:32
There are no external cameras.
30
80520
1616
01:34
Each flying machine uses onboard sensors
to determine its location in space
to determine its location in space
31
82160
5056
01:39
and onboard computation
to determine what its actions should be.
to determine what its actions should be.
32
87240
3440
01:43
The only external commands
are high-level ones
are high-level ones
33
91280
2176
01:45
such as "take off" and "land."
34
93480
1520
02:10
This is a so-called tail-sitter.
35
118560
1896
02:12
It's an aircraft that tries
to have its cake and eat it.
to have its cake and eat it.
36
120480
3456
02:15
Like other fixed-wing aircraft,
it is efficient in forward flight,
it is efficient in forward flight,
37
123960
3256
02:19
much more so than helicopters
and variations thereof.
and variations thereof.
38
127240
3215
02:22
Unlike most other
fixed-wing aircraft, however,
fixed-wing aircraft, however,
39
130479
2417
02:24
it is capable of hovering,
40
132920
1496
02:26
which has huge advantages
for takeoff, landing
for takeoff, landing
41
134440
3136
02:29
and general versatility.
42
137600
1240
02:31
There is no free lunch, unfortunately.
43
139480
2096
02:33
One of the limitations with tail-sitters
44
141600
1936
02:35
is that they're susceptible
to disturbances such as wind gusts.
to disturbances such as wind gusts.
45
143560
2976
02:38
We're developing new control
architectures and algorithms
architectures and algorithms
46
146560
2696
02:41
that address this limitation.
47
149280
1400
02:50
The idea is for the aircraft to recover
48
158560
2256
02:52
no matter what state it finds itself in,
49
160840
2640
03:03
and through practice,
improve its performance over time.
improve its performance over time.
50
171400
2920
03:15
(Applause)
51
183600
4400
03:22
OK.
52
190000
1200
03:33
When doing research,
53
201480
1216
03:34
we often ask ourselves
fundamental abstract questions
fundamental abstract questions
54
202720
2976
03:37
that try to get at the heart of a matter.
55
205720
2560
03:41
For example, one such question would be,
56
209240
2216
03:43
what is the minimum number of moving parts
needed for controlled flight?
needed for controlled flight?
57
211480
4016
03:47
Now, there are practical reasons
58
215520
1576
03:49
why you may want to know
the answer to such a question.
the answer to such a question.
59
217120
2616
03:51
Helicopters, for example,
60
219760
1256
03:53
are affectionately known
as machines with a thousand moving parts
as machines with a thousand moving parts
61
221040
3856
03:56
all conspiring to do you bodily harm.
62
224920
2680
04:00
It turns out that decades ago,
63
228640
1856
04:02
skilled pilots were able to fly
remote-controlled aircraft
remote-controlled aircraft
64
230520
3176
04:05
that had only two moving parts:
65
233720
1896
04:07
a propeller and a tail rudder.
66
235640
1840
04:10
We recently discovered
that it could be done with just one.
that it could be done with just one.
67
238000
3120
04:13
This is the monospinner,
68
241600
1256
04:14
the world's mechanically simplest
controllable flying machine,
controllable flying machine,
69
242880
3256
04:18
invented just a few months ago.
70
246160
1776
04:19
It has only one moving part, a propeller.
71
247960
3136
04:23
It has no flaps, no hinges, no ailerons,
72
251120
3576
04:26
no other actuators,
no other control surfaces,
no other control surfaces,
73
254720
2535
04:29
just a simple propeller.
74
257279
1281
04:31
Even though it's mechanically simple,
75
259320
2055
04:33
there's a lot going on
in its little electronic brain
in its little electronic brain
76
261399
2497
04:35
to allow it to fly in a stable fashion
and to move anywhere it wants in space.
and to move anywhere it wants in space.
77
263920
4216
04:40
Even so, it doesn't yet have
78
268160
1456
04:41
the sophisticated algorithms
of the tail-sitter,
of the tail-sitter,
79
269640
2336
04:44
which means that in order
to get it to fly,
to get it to fly,
80
272000
2000
04:46
I have to throw it just right.
81
274024
1720
04:48
And because the probability
of me throwing it just right is very low,
of me throwing it just right is very low,
82
276800
4016
04:52
given everybody watching me,
83
280840
1896
04:54
what we're going to do instead
84
282760
1456
04:56
is show you a video
that we shot last night.
that we shot last night.
85
284240
2176
04:58
(Laughter)
86
286440
1560
05:10
(Applause)
87
298320
3320
05:23
If the monospinner
is an exercise in frugality,
is an exercise in frugality,
88
311080
3736
05:26
this machine here, the omnicopter,
with its eight propellers,
with its eight propellers,
89
314840
3456
05:30
is an exercise in excess.
90
318320
2080
05:32
What can you do with all this surplus?
91
320920
2136
05:35
The thing to notice
is that it is highly symmetric.
is that it is highly symmetric.
92
323080
2381
05:37
As a result, it is ambivalent
to orientation.
to orientation.
93
325920
2856
05:40
This gives it an extraordinary capability.
94
328800
2696
05:43
It can move anywhere it wants in space
95
331520
2216
05:45
irrespective of where it is facing
96
333760
2736
05:48
and even of how it is rotating.
97
336520
1800
05:51
It has its own complexities,
98
339040
1376
05:52
mainly having to do
with the interacting flows
with the interacting flows
99
340440
2656
05:55
from its eight propellers.
100
343120
1240
05:56
Some of this can be modeled,
while the rest can be learned on the fly.
while the rest can be learned on the fly.
101
344680
3856
06:00
Let's take a look.
102
348560
1200
06:44
(Applause)
103
392760
3760
06:52
If flying machines are going
to enter part of our daily lives,
to enter part of our daily lives,
104
400720
2905
06:55
they will need to become
extremely safe and reliable.
extremely safe and reliable.
105
403649
2477
06:58
This machine over here
106
406760
1376
07:00
is actually two separate
two-propeller flying machines.
two-propeller flying machines.
107
408160
3136
07:03
This one wants to spin clockwise.
108
411320
1736
07:05
This other one wants
to spin counterclockwise.
to spin counterclockwise.
109
413080
2496
07:07
When you put them together,
110
415600
1336
07:08
they behave like one
high-performance quadrocopter.
high-performance quadrocopter.
111
416960
2600
07:23
If anything goes wrong, however --
112
431640
1656
07:25
a motor fails, a propeller fails,
electronics, even a battery pack --
electronics, even a battery pack --
113
433320
4456
07:29
the machine can still fly,
albeit in a degraded fashion.
albeit in a degraded fashion.
114
437800
3336
07:33
We're going to demonstrate this to you now
by disabling one of its halves.
by disabling one of its halves.
115
441160
3520
07:56
(Applause)
116
464240
3160
08:03
This last demonstration
117
471320
1336
08:04
is an exploration of synthetic swarms.
118
472680
2440
08:07
The large number of autonomous,
coordinated entities
coordinated entities
119
475560
2976
08:10
offers a new palette
for aesthetic expression.
for aesthetic expression.
120
478560
2736
08:13
We've taken commercially available
micro quadcopters,
micro quadcopters,
121
481320
2976
08:16
each weighing less
than a slice of bread, by the way,
than a slice of bread, by the way,
122
484320
2496
08:18
and outfitted them
with our localization technology
with our localization technology
123
486840
2416
08:21
and custom algorithms.
124
489280
1576
08:22
Because each unit
knows where it is in space
knows where it is in space
125
490880
2096
08:25
and is self-controlled,
126
493000
1376
08:26
there is really no limit to their number.
127
494400
2200
08:55
(Applause)
128
523840
3440
09:19
(Applause)
129
547360
4480
10:18
(Applause)
130
606040
3640
10:35
Hopefully, these demonstrations
will motivate you to dream up
will motivate you to dream up
131
623960
3096
10:39
new revolutionary roles
for flying machines.
for flying machines.
132
627080
2400
10:42
That ultrasafe one over there for example
133
630560
2096
10:44
has aspirations to become
a flying lampshade on Broadway.
a flying lampshade on Broadway.
134
632680
3016
10:47
(Laughter)
135
635720
2056
10:49
The reality is that it is
difficult to predict
difficult to predict
136
637800
2176
10:52
the impact of nascent technology.
137
640000
1640
10:54
And for folks like us, the real reward
is the journey and the act of creation.
is the journey and the act of creation.
138
642120
4856
10:59
It's a continual reminder
139
647000
1416
11:00
of how wonderful and magical
the universe we live in is,
the universe we live in is,
140
648440
2920
11:03
that it allows creative, clever creatures
141
651840
2936
11:06
to sculpt it in such spectacular ways.
142
654800
2520
11:09
The fact that this technology
143
657800
1776
11:11
has such huge commercial
and economic potential
and economic potential
144
659600
3416
11:15
is just icing on the cake.
145
663040
1416
11:16
Thank you.
146
664480
1216
11:17
(Applause)
147
665720
3240
ABOUT THE SPEAKER
Raffaello D'Andrea - Autonomous systems pioneerRaffaello D'Andrea explores the possibilities of autonomous technology by collaborating with artists, engineers and entrepreneurs.
Why you should listen
Raffaello D'Andrea combines academics, business, and the arts to explore the capabilities of autonomous systems. As part of his research as professor of dynamic systems and control at the Swiss Federal Institute of Technology (ETH Zürich), he and his collaborators enchant viewers with works like the self-destructing, self-assembling Robotic Chair, or the Balancing Cube that can perch itself on its corners.
D’Andrea and his team created the Flying Machine Arena to test the gravity-defying abilities of their athletic flying robots. Building on research in the Flying Machine Arena, ETH Zürich partnered with its spin-off company Verity Studios and with Cirque du Soleil to create “Sparked,” a short film showcasing the unexpected airborne dexterity of quadcopters. D’Andrea is the co-founder of Kiva Systems, a robotics company that develops intelligent automated warehouse systems and that was acquired by Amazon in 2012.
Raffaello D'Andrea | Speaker | TED.com