The astounding athletic power of quadcopters | Raffaello D'Andrea

10,776,689 views ใƒป 2013-06-12

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

00:00
Translator: Joseph Geni Reviewer: Morton Bast
0
0
7000
ืžืชืจื’ื: Yubal Masalker ืžื‘ืงืจ: Ido Dekkers
00:23
So what does it mean for a machine to be athletic?
1
23544
2523
ืื– ืžื” ื–ื” ืื•ืžืจ ืฉืžื›ื•ื ื” ื”ื™ื ืืชืœื˜ื™ืช?
00:27
We will demonstrate the concept of machine athleticism
2
27059
2825
ื ืฆื™ื’ ื›ืืŸ ืืช ื”ืชืคื™ืกื” ืœื’ื‘ื™ ืืชืœื˜ื™ื•ืช ืฉืœ ืžื›ื•ื ื”
00:29
and the research to achieve it
3
29908
1696
ื•ืืช ืขื‘ื•ื“ืช ื”ืžื—ืงืจ ื›ื“ื™ ืœื”ืฉื™ื’ื”
00:31
with the help of these flying machines called quadrocopters,
4
31628
2858
ื‘ืขื–ืจืช ื”ืžื›ื•ื ื•ืช ื”ืžืขื•ืคืคื•ืช ื”ืœืœื• ื”ื ืงืจืื•ืช ืžืกื•ืงื™-ืงื•ืื“,
00:34
or quads, for short.
5
34510
1047
ืื• ื‘ืงื™ืฆื•ืจ, ืงื•ืื“.
00:38
Quads have been around for a long time.
6
38291
1893
ื”ืžืกื•ืงื™ื ื ืžืฆืื™ื ืื™ืชื ื• ื›ื‘ืจ ื–ืžืŸ ืจื‘.
00:40
They're so popular these days because they're mechanically simple.
7
40501
3277
ื”ืกื™ื‘ื” ืฉื”ื ื›ื” ืคื•ืคื•ืœืจื™ื ื‘ื™ืžื™ื ื• ื”ื™ื ื”ื™ื•ืชืŸ ืคืฉื•ื˜ื™ื ืžื‘ื—ื™ื ื” ืžื™ื›ื ื™ืช.
00:44
By controlling the speeds of these four propellers,
8
44255
2389
ื‘ืืžืฆืขื•ืช ืฉืœื™ื˜ื” ืขืœ ืžื”ื™ืจื•ืช ื”ืกื™ื‘ื•ื‘ ืืจื‘ืขืช ื”ืจื•ื˜ื•ืจื™ื ื”ืœืœื•,
00:46
these machines can roll, pitch, yaw,
9
46668
2745
ื”ื›ืœื™ื ื™ื›ื•ืœื™ื ืœื”ืชื’ืœื’ืœ, ืœื”ื™ื˜ื•ืช ื•ืœืขืœืจื“,
00:49
and accelerate along their common orientation.
10
49437
2171
ืœืฉื ื•ืช ื›ื™ื•ื•ืŸ ื•ืœื”ืื™ืฅ ื‘ื›ื™ื•ื•ืŸ ืื—ื“ ืžืฉื•ืชืฃ ืœื›ื•ืœื.
00:52
On board are also a battery, a computer,
11
52179
3198
ื”ื ื ื•ืฉืื™ื ืขืœื™ื”ื ืกื•ืœืœื”, ืžื—ืฉื‘,
00:55
various sensors and wireless radios.
12
55401
1953
ื—ื™ื™ืฉื ื™ื ืฉื•ื ื™ื ื•ืจื“ื™ื• ืืœื—ื•ื˜ื™.
00:59
Quads are extremely agile, but this agility comes at a cost.
13
59298
3309
ื”ืžืกื•ืงื™ื ืžืื•ื“ ืงืœื™ ืชื ื•ืขื”, ืื‘ืœ ืงืœื•ืช ื–ื• ื’ื•ื‘ื” ืžื—ื™ืจ.
01:03
They are inherently unstable,
14
63762
2088
ื”ื ืœื ื™ืฆื™ื‘ื™ื ื‘ื™ืกื•ื“ื
01:05
and they need some form of automatic feedback control
15
65874
2571
ื•ื”ื ื–ืงื•ืงื™ื ืœืื™ื–ื” ืฉื”ื•ื ืžืฉื•ื‘ ืขื ื‘ืงืจื” ืื•ื˜ื•ืžื˜ื™ืช
01:08
in order to be able to fly.
16
68469
1373
ื›ื“ื™ ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื ืœื˜ื•ืก.
01:16
So, how did it just do that?
17
76259
2253
ืื– ื›ื™ืฆื“ ืขื•ืฉื™ื ื–ืืช?
01:19
Cameras on the ceiling and a laptop
18
79497
1794
ืžืฆืœืžื•ืช ื‘ืชืงืจื” ื•ืžื—ืฉื‘ ื ื™ื™ื“
01:21
serve as an indoor global positioning system.
19
81315
2430
ืžืฉืžืฉื™ื ื›ืžืขืจื›ืช ืžื™ืงื•ื ื’ืœื•ื‘ืœื™ืช ื‘ืžืชื—ื ืกื’ื•ืจ.
01:24
It's used to locate objects in the space
20
84301
2039
ื”ื™ื ืžืฉืžืฉืช ืœืื™ืชื•ืจ ืขืฆืžื™ื ื‘ืžืจื—ื‘
01:26
that have these reflective markers on them.
21
86364
2088
ืฉื ื•ืฉืื™ื ืขืœื™ื”ื ืกืžื ื™ื ืžื—ื–ื™ืจื™ื ืืœื”.
01:28
This data is then sent to another laptop
22
88476
2038
ื ืชื•ื ื™ื ื”ืœืœื• ืžืฉื•ื’ืจื™ื ืœืžื—ืฉื‘ ื ื™ื™ื“ ืื—ืจ
01:30
that is running estimation and control algorithms,
23
90538
2358
ืฉืžืจื™ืฅ ืืœื’ื•ืจื™ืชืžื™ื ืฉืœ ืื•ืžื“ืŸ ื•ื‘ืงืจื”,
01:32
which in turn sends commands to the quad,
24
92920
1954
ืฉื‘ืชื•ืจื• ืฉื•ืœื— ืคืงื•ื“ื•ืช ืœื›ืœื™-ื”ื˜ื™ืก,
01:34
which is also running estimation and control algorithms.
25
94898
2633
ืฉืืฃ ื”ื•ื ืžืจื™ืฅ ืืœื’ื•ืจื™ืชืžื™ ืื•ืžื“ืŸ ื•ื‘ืงืจื”.
01:41
The bulk of our research is algorithms.
26
101864
2242
ื”ื ืชื— ื”ืขื™ืงืจื™ ืฉืœ ื”ืžื—ืงืจ ืฉืœื ื• ื”ื•ื ืืœื’ื•ืจื™ืชืžื™ื.
01:44
It's the magic that brings these machines to life.
27
104130
2576
ื”ื ื”ืงืกื ืฉืžืขื•ืจืจ ืœื—ื™ื™ื ืืช ื”ื›ืœื™ื ื”ืœืœื•.
01:48
So how does one design the algorithms that create a machine athlete?
28
108291
4033
ืื– ื›ื™ืฆื“ ื‘ื•ื ื™ื ืืช ื”ืืœื’ื•ืจื™ืชืžื™ื ืฉื™ื•ืฆืจื™ื ืืช ื”ืืชืœื˜ ื”ืžื›ื ื™?
01:52
We use something broadly called model-based design.
29
112792
2587
ืื ื• ืžืฉืชืžืฉื™ื ื‘ืžื” ืฉื ืงืจื ื‘ื’ื“ื•ืœ ืชื›ื ื•ืŸ ืžื‘ื•ืกืก-ืžื•ื“ืœื™ื.
01:55
We first capture the physics
30
115839
1366
ืชื—ื™ืœื” ืื ื• ืžื ืชื—ื™ื ืืช ื”ืคื™ื–ื™ืงื”
01:57
with a mathematical model of how the machines behave.
31
117229
2983
ื‘ืขื–ืจืช ืžื•ื“ืœ ืžืชืžื˜ื™ ืฉืœ ืื™ืš ื”ืžื›ื•ื ื” ืžืชื ื”ื’ืช.
02:00
We then use a branch of mathematics called control theory
32
120636
3950
ืœืื—ืจ-ืžื›ืŸ ืื ื• ืžืฉืชืžืฉื™ื ื‘ืขื ืฃ ืžืชืžื˜ื™ ืฉื ืงืจื ืชืื•ืจื™ื™ืช ื‘ืงืจื”
02:04
to analyze these models
33
124610
1657
ื›ื“ื™ ืœื ืชื— ืžื•ื“ืœื™ื ืืœื”
02:06
and also to synthesize algorithms for controlling them.
34
126291
2770
ื•ื’ื ื›ื“ื™ ืœืžื–ื’ ืืœื’ื•ืจื™ืชืžื™ื ื‘ืฉื‘ื™ืœ ืœืฉืœื•ื˜ ื‘ื”ื.
02:09
For example, that's how we can make the quad hover.
35
129820
3024
ืœื“ื•ื’ืžื, ื›ืš ืื ื• ื’ื•ืจืžื™ื ืœื›ืœื™ ืœืจื—ืฃ.
02:12
We first captured the dynamics with a set of differential equations.
36
132868
3290
ืชื—ื™ืœื” ื”ื›ืจื ื• ืืช ื”ื›ื•ื—ื•ืช ื•ื”ืชื ื•ืขื•ืช ื‘ืขื–ืจืช ืžืขืจื›ืช ืžืฉื•ื•ืื•ืช ื“ื™ืคืจื ืฆื™ืืœื™ื•ืช.
02:16
We then manipulate these equations with the help of control theory
37
136182
3452
ืœืื—ืจ-ืžื›ืŸ ืฉื™ื ื™ื ื• ืžืฉื•ื•ืื•ืช ืืœื• ื‘ืขื–ืจืช ืชืื•ืจื™ื™ืช ื‘ืงืจื”
02:19
to create algorithms that stabilize the quad.
38
139658
2937
ื›ื“ื™ ืœื™ืฆื•ืจ ืืœื’ื•ืจื™ืชืžื™ื ืืฉืจ ืžื™ื™ืฆื‘ื™ื ืืช ื”ื›ืœื™.
02:23
Let me demonstrate the strength of this approach.
39
143130
2293
ืืฆื™ื’ ืขื›ืฉื™ื• ืืช ื™ืชืจื•ื ื•ืช ื”ื’ื™ืฉื” ื”ื–ื•.
02:29
Suppose that we want this quad to not only hover
40
149198
2633
ื ื ื™ื— ืฉืื ื• ืจื•ืฆื™ื ืฉื”ื›ืœื™ ื”ื–ื” ืœื ืจืง ื™ืจื—ืฃ
02:31
but to also balance this pole.
41
151855
1659
ืืœื ื’ื ื™ื—ื–ื™ืง ืžื•ื˜ ื–ื” ื‘ืฉื™ื•ื•ื™-ืžืฉืงืœ.
02:34
With a little bit of practice,
42
154520
1774
ืขื ืงืฆืช ืชืจื’ื•ืœ,
02:36
it's pretty straightforward for a human being to do this,
43
156318
2683
ื–ื” ื“ื™ ืงืœ ืœืื“ื ืœื‘ืฆืข ื–ืืช,
02:39
although we do have the advantage of having two feet on the ground
44
159025
3191
ื’ื ื‘ื’ื™ืŸ ื”ื™ืชืจื•ืŸ ืฉืœ ืฉืชื™ ืจื’ืœื™ื ื• ืขืœ ื”ืงืจืงืข
02:42
and the use of our very versatile hands.
45
162240
2059
ื•ื‘ื’ื™ืŸ ื”ืฉื™ืžื•ืฉ ื‘ื™ื“ื™ื ื• ื‘ืขืœื•ืช ื”ื™ื›ื•ืœื•ืช ื”ืจื‘-ื’ื•ื ื™ื•ืช.
02:44
It becomes a little bit more difficult
46
164804
2738
ื–ื” ื”ื•ืคืš ืœืงืฆืช ื™ื•ืชืจ ืงืฉื”
02:47
when I only have one foot on the ground
47
167566
1979
ื›ืืฉืจ ื™ืฉ ืœื™ ืจืง ืจื’ืœ ืื—ืช ืขืœ ื”ืงืจืงืข
02:49
and when I don't use my hands.
48
169569
1602
ื•ืื™ื ื™ ืžืฉืชืžืฉ ื‘ื™ื“ื™ื™.
02:52
Notice how this pole has a reflective marker on top,
49
172733
2562
ืฉื™ืžื• ืœื‘ ืฉืœืžื•ื˜ ื–ื” ืžื•ืฆืžื“ ืกืžืŸ ืœืžืขืœื”,
02:55
which means that it can be located in the space.
50
175319
2661
ืฉื–ื” ืื•ืžืจ ืฉื ื™ืชืŸ ืœืืชืจ ืืช ืžื™ืงื•ืžื• ื‘ืžืจื—ื‘.
03:04
(Audience) Oh!
51
184276
1067
(ืงื”ืœ) ืื•ื”!
03:05
(Applause)
52
185367
1259
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
03:09
(Applause ends)
53
189695
1668
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื ืžืกืชื™ื™ืžื•ืช)
03:11
You can notice that this quad is making fine adjustments
54
191387
2641
ื ื™ืชืŸ ืœืจืื•ืช ืฉื›ืœื™ ื–ื” ืขื•ืฉื” ืชื™ืงื•ื ื™ื ืงืœื™ื
03:14
to keep the pole balanced.
55
194052
1374
ื›ื“ื™ ืœืฉืžื•ืจ ืขืœ ืื™ื–ื•ืŸ ื”ืžื•ื˜.
03:16
How did we design the algorithms to do this?
56
196417
2072
ื›ื™ืฆื“ ื‘ื ื™ื ื• ืืช ื”ืืœื’ื•ืจื™ืชืžื™ื ื”ืขื•ืฉื™ื ื–ืืช?
03:19
We added the mathematical model of the pole
57
199027
2081
ื”ื•ืกืคื ื• ืืช ื”ืžื•ื“ืœ ื”ืžืชืžื˜ื™ ืฉืœ ื”ืžื•ื˜
03:21
to that of the quad.
58
201132
1447
ืœื–ื” ืฉืœ ื”ืžืกื•ืง.
03:22
Once we have a model of the combined quad-pole system,
59
202603
2978
ื‘ืจื’ืข ืฉื™ืฉ ืืช ื”ืžื•ื“ืœ ื”ืžืฉื•ืœื‘ ืฉืœ ืžืขืจื›ืช ืžืกื•ืง-ืžื•ื˜,
03:25
we can use control theory to create algorithms for controlling it.
60
205605
3319
ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ืชื™ืื•ืจื™ื™ืช ื‘ืงืจื” ื›ื“ื™ ืœื™ืฆื•ืจ ืืœื’ื•ืจื™ืชืžื™ื ืฉื™ืฉืœื˜ื• ื‘ื”.
03:30
Here, you see that it's stable,
61
210538
1564
ื”ื ื”, ื ื™ืชืŸ ืœืจืื•ืช ืฉื”ื•ื ื™ืฆื™ื‘,
03:32
and even if I give it little nudges,
62
212126
2669
ื•ืœืžืจื•ืช ืื ื™ ื ื•ืชืŸ ืœื• ืžื›ื•ืช ืงืœื•ืช,
03:34
it goes back --
63
214819
1166
ื”ื•ื ื—ื•ื–ืจ
03:37
to the nice, balanced position.
64
217589
1769
ืœืžืฆื‘ ืฉืœ ืฉื™ื•ื•ื™-ืžืฉืงืœ.
03:40
We can also augment the model
65
220087
1390
ืื ื• ื’ื ื™ื›ื•ืœื™ื ืœื”ืจื—ื™ื‘ ืืช ื”ืžื•ื“ืœ
03:41
to include where we want the quad to be in space.
66
221501
2371
ืฉื™ื›ืœื•ืœ ืืช ืžื™ืงื•ืžื• ื”ืจืฆื•ื™ ืœื ื• ืฉืœ ื”ื›ืœื™ ื‘ืžืจื—ื‘.
03:44
Using this pointer, made out of reflective markers,
67
224428
2943
ื‘ืขื–ืจืช ืžืฆื‘ื™ืข ื–ื” ืฉืขืฉื•ื™ ืžืกืžื ื™ื ืžื—ื–ื™ืจื™ื,
03:47
I can point to where I want the quad to be in space
68
227395
2429
ืืคืฉืจ ืœื”ืฆื‘ื™ืข ืขืœ ื”ืžื™ืงื•ื ื”ืจืฆื•ื™ ืœื ื• ืฉืœ ื”ื›ืœื™ ื‘ืžืจื—ื‘,
03:49
a fixed distance away from me.
69
229848
1597
ื‘ืžืจื—ืง ื”ืงื‘ื•ืข ืžืจืืฉ ืžืžื ื™.
03:59
(Laughter)
70
239785
1880
(ืฆื—ื•ืง)
04:07
The key to these acrobatic maneuvers is algorithms,
71
247763
3126
ื”ืžืคืชื— ืœืชืžืจื•ื ื™ื ื”ืืงืจื•ื‘ื˜ื™ื™ื ื”ืœืœื• ื”ื ืืœื’ื•ืจื™ืชืžื™ื,
04:10
designed with the help of mathematical models
72
250913
2293
ืฉื ื‘ื ื• ื‘ืขื–ืจืช ืžื•ื“ืœื™ื ืžืชืžื˜ื™ื™ื
04:13
and control theory.
73
253230
1185
ื•ืชืื•ืจื™ื™ืช ื‘ืงืจื”.
04:15
Let's tell the quad to come back here
74
255246
1985
ื›ืขืช ื ืืžืจ ืœืžืกื•ืง ืœื—ื–ื•ืจ ืœื›ืืŸ
04:17
and let the pole drop,
75
257255
1664
ื•ื ืชืŸ ืœืžื•ื˜ ืœื™ืคื•ืœ.
04:18
and I will next demonstrate the importance
76
258943
2131
ืืชืืจ ื›ืขืช ืืช ื”ื—ืฉื™ื‘ื•ืช
04:21
of understanding physical models
77
261098
1989
ืฉื‘ื”ื‘ื ืช ื”ืžื•ื“ืœื™ื ื”ืคื™ื–ื™ืงืœื™ื™ื
04:23
and the workings of the physical world.
78
263111
2045
ื•ืฉืœ ื“ืจืš-ื”ืคืขื•ืœื” ืฉืœ ื”ืขื•ืœื ื”ืคื™ื–ื™.
04:37
Notice how the quad lost altitude when I put this glass of water on it.
79
277462
3828
ืฉื™ืžื• ืœื‘ ืื™ืš ื”ื›ืœื™ ืื™ื‘ื“ ื’ื•ื‘ื” ื›ืฉื”ื ื—ืชื™ ืขืœื™ื• ื›ื•ืก ืžื™ื.
04:41
Unlike the balancing pole,
80
281314
1403
ืฉืœื ื›ืžื• ื”ืžื•ื˜ ื”ืžืื•ื–ืŸ,
04:42
I did not include the mathematical model of the glass
81
282741
3015
ืœื ื”ื›ืœืœืชื™ ืืช ื”ืžื•ื“ืœ ื”ืžืชืžื˜ื™ ืฉืœ ื”ื›ื•ืก.
04:45
in the system.
82
285780
1307
ื‘ืžืขืจื›ืช
04:47
In fact, the system doesn't even know that the glass is there.
83
287111
3013
ืœืžืขืฉื”, ื”ืžืขืจื›ืช ืืคื™ืœื• ืื™ื ื” ื™ื•ื“ืขืช ืฉื›ื•ืก ื”ืžื™ื ื ืžืฆืืช ืขืœื™ื”.
04:50
Like before, I could use the pointer to tell the quad
84
290148
2738
ื›ืžื• ืžืงื•ื“ื, ื™ื›ื•ืœืชื™ ืœื”ืฉืชืžืฉ ื‘ืžืฆื‘ื™ืข ื›ื“ื™ ืœื•ืžืจ ืœืžืกื•ืง
04:52
where I want it to be in space.
85
292910
1571
ื”ื™ื›ืŸ ืื ื™ ืจื•ืฆื” ืฉื”ื•ื ื™ืฉื”ื” ื‘ืžืจื—ื‘.
04:57
(Applause)
86
297684
1799
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
05:03
(Applause ends)
87
303531
1540
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื ืžืกืชื™ื™ืžื•ืช)
05:05
Okay, you should be asking yourself,
88
305515
1739
ืืชื ืฆืจื™ื›ื™ื ืœืฉืื•ืœ ืืช ืขืฆืžื›ื,
05:07
why doesn't the water fall out of the glass?
89
307278
2062
ื›ื™ืฆื“ ื”ืžื™ื ืœื ื ืฉืคื›ื™ื ืžื”ื›ื•ืก?
05:09
Two facts.
90
309888
1030
ืฉืชื™ ืขื•ื‘ื“ื•ืช:
05:10
The first is that gravity acts on all objects in the same way.
91
310942
3633
ื”ืจืืฉื•ื ื” ื”ื™ื ืฉื›ื•ื—-ื”ืžืฉื™ื›ื” ืคื•ืขืœ ืขืœ ื›ืœ ื”ืขืฆืžื™ื ื‘ืื•ืชื• ืื•ืคืŸ.
05:15
The second is that the propellers
92
315051
1779
ื”ืฉื ื™ื™ื” ื”ื•ื ืฉื”ืจื•ื˜ื•ืจื™ื ื›ื•ืœื ืžืฆื‘ื™ืขื™ื
05:16
are all pointing in the same direction of the glass, pointing up.
93
316854
3073
ื‘ืื•ืชื• ื›ื™ื•ื•ืŸ ืฉืœ ื”ื›ื•ืก, ืžื›ื•ื•ื ื™ื ืžืขืœื”.
05:20
You put these two things together,
94
320642
1647
ื›ืืฉืจ ืžื—ื‘ืจื™ื ืืช ืฉื ื™ ื”ื“ื‘ืจื™ื, ื”ืชื•ืฆืื” ื”ืกื•ืคื™ืช ื”ื™ื
05:22
the net result is that all side forces on the glass are small
95
322313
3555
ืฉื›ืœ ื”ื›ื•ื—ื•ืช ื”ืžืฉื ื™ื™ื ืขืœ ื”ื›ื•ืก ื”ื ืงื˜ื ื™ื
05:25
and are mainly dominated by aerodynamic effects,
96
325892
2277
ื•ืžื•ืฉืคืขื™ื ื‘ืขื™ืงืจ ืžืืคืงื˜ื™ื ืื•ื•ื™ืจื•ื“ื™ื ืžื™ื™ื,
05:28
which at these speeds are negligible.
97
328193
1961
ืฉื‘ืžื”ื™ืจื•ื™ื•ืช ืืœื• ื”ื ื–ื ื™ื—ื™ื.
05:35
And that's why you don't need to model the glass.
98
335336
2334
ื•ื–ื• ื”ืกื™ื‘ื” ืžื“ื•ืข ืื™ืŸ ืฆื•ืจืš ืœื‘ื ื•ืช ืžื•ื“ืœ ืœื›ื•ืก.
05:37
It naturally doesn't spill, no matter what the quad does.
99
337694
2741
ื‘ืื•ืคืŸ ื˜ื‘ืขื™ ื”ืžื™ื ืœื ื ืฉืคื›ื™ื ื•ืœื ืžืฉื ื” ืžื” ื”ืžืกื•ืง ืขื•ืฉื”.
05:44
(Audience) Oh!
100
344500
2563
(ืงื”ืœ) ืื•ื”!
05:50
(Applause)
101
350340
1858
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
05:55
(Applause ends)
102
355159
1255
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื ืžืกืชื™ื™ืžื•ืช)
05:57
The lesson here
103
357940
1114
ื”ืœืงื— ื›ืืŸ
05:59
is that some high-performance tasks are easier than others,
104
359078
4013
ื”ื•ื ืฉื—ืœืง ืžื”ืžืฉื™ืžื•ืช ื”ื“ื•ืจืฉื•ืช ื‘ื™ืฆื•ืขื™ื ื’ื‘ื•ื”ื™ื ื”ืŸ ืงืœื•ืช ื™ื•ืชืจ ืžืื—ืจื•ืช,
06:03
and that understanding the physics of the problem
105
363115
2337
ื•ืฉื”ื”ื‘ื ื” ื”ืคื™ื–ื™ืงืœื™ืช ืฉืœ ื”ื‘ืขื™ื”
06:05
tells you which ones are easy and which ones are hard.
106
365476
2694
ืื•ืžืจืช ืœื ื• ืืœื• ื‘ืขื™ื•ืช ื”ืŸ ืงืœื•ืช ื•ืืœื• ืงืฉื•ืช.
06:08
In this instance, carrying a glass of water is easy.
107
368194
2510
ื‘ืžืงืจื” ื–ื”, ืœืฉืืช ื›ื•ืก ืžื™ื ื–ื” ืงืœ.
06:10
Balancing a pole is hard.
108
370728
1630
ืœืฉืžื•ืจ ืขืœ ืฉื™ื•ื•ื™-ืžืฉืงืœ ืฉืœ ืžื•ื˜ ื–ื” ืงืฉื”.
06:14
We've all heard stories of athletes performing feats while physically injured.
109
374171
4271
ื›ื•ืœื ื• ืฉืžืขื ื• ืกื™ืคื•ืจื™ื ืขืœ ืกืคื•ืจื˜ืื™ื ืฉืžื’ื™ืขื™ื ืœื”ื™ืฉื’ื™ื ื’ื‘ื•ื”ื™ื ื‘ืขื•ื“ื ืคืฆื•ืขื™ื.
06:18
Can a machine also perform with extreme physical damage?
110
378466
3185
ื”ืื ื’ื ืžื›ื•ื ื” ืžืกื•ื’ืœืช ืœื‘ื™ืฆื•ืขื™ื ื›ืืฉืจ ื™ืฉ ืœื” ื ื–ืง ื—ืžื•ืจ?
06:22
Conventional wisdom says
111
382636
1485
ื”ื“ืขื” ื”ืžืงื•ื‘ืœืช ื”ื™ื
06:24
that you need at least four fixed motor propeller pairs in order to fly,
112
384145
3883
ืฉืฆืจื™ืš ืœืคื—ื•ืช 4 ื–ื•ื’ื•ืช ืฉืœ ืจื•ื˜ื•ืจื™ื ื›ื“ื™ ืœื˜ื•ืก,
06:28
because there are four degrees of freedom to control:
113
388052
2524
ืžื›ื™ื•ื•ืŸ ืฉื™ืฉ 4 ื“ืจื’ื•ืช ื—ื•ืคืฉ ืฉืฆืจื™ืš ืœืฉืœื•ื˜ ื‘ื”ืŸ:
06:30
roll, pitch, yaw and acceleration.
114
390600
1965
ื’ืœื’ื•ืœ, ืขืœืจื•ื“, ืกื™ื‘ืกื•ื‘ ื•ืชืื•ืฆื”.
06:33
Hexacopters and octocopters, with six and eight propellers,
115
393192
3221
ืœืžืกื•ืงื™ื ื‘ืขืœื™ 6 ื•-8 ืจื•ื˜ื•ืจื™ื,
06:36
can provide redundancy,
116
396437
1459
ื™ืฉ ื™ืชื™ืจื•ืช,
06:37
but quadrocopters are much more popular
117
397920
1965
ืื‘ืœ ืžืกื•ืงื™-ืจื‘ื™ืขื™ื™ื” ื”ืจื‘ื” ื™ื•ืชืจ ื ืคื•ืฆื™ื
06:39
because they have the minimum number of fixed motor propeller pairs: four.
118
399909
3686
ื›ื™ ื™ืฉ ืœื”ื ืืช ื”ืžืกืคืจ ื”ืžื™ื ื™ืžืœื™ ืฉืœ ืจื•ื˜ื•ืจื™ื: 4
06:44
Or do they?
119
404168
1016
ื”ืืžื ื?
06:52
(Audience) Oh!
120
412827
1000
(ืงื”ืœ) ืืŸื”!
06:54
(Laughter)
121
414407
1000
(ืฆื—ื•ืง)
07:01
If we analyze the mathematical model of this machine
122
421334
2718
ื›ืืฉืจ ืžื ืชื—ื™ื ืืช ื”ืžื•ื“ืœ ื”ืžืชืžื˜ื™ ืฉืœ ืžื›ื•ื ื” ื–ื•
07:04
with only two working propellers,
123
424076
2152
ืขื ืฉื ื™ ืจื•ื˜ื•ืจื™ื ืคื•ืขืœื™ื ื‘ืœื‘ื“,
07:06
we discover that there's an unconventional way to fly it.
124
426252
2740
ืžื’ืœื™ื ืฉื™ืฉ ื“ืจืš ื™ื•ืฆืืช-ื“ื•ืคืŸ ืœื”ื˜ื™ืกื”.
07:19
We relinquish control of yaw,
125
439980
1682
ืžื•ื•ืชืจื™ื ืขืœ ื”ืฉืœื™ื˜ื” ื‘ืกื™ื‘ืกื•ื‘,
07:21
but roll, pitch and acceleration can still be controlled
126
441686
3046
ืื‘ืœ ืขื“ื™ื™ืŸ ื ื™ืชืŸ ืœืฉืœื•ื˜ ื‘ื’ืœื’ื•ืœ, ืขืœืจื•ื“ ื•ืชืื•ืฆื”
07:24
with algorithms that exploit this new configuration.
127
444756
3346
ื‘ืืžืฆืขื•ืช ืืœื’ื•ืจื™ืชืžื™ื ื”ืžื ืฆืœื™ื ืชืฆื•ืจื” ื—ื“ืฉื” ื–ื•.
07:33
Mathematical models tell us exactly when and why this is possible.
128
453668
4289
ืžื•ื“ืœื™ื ืžืชืžื˜ื™ื™ื ืื•ืžืจื™ื ืœื ื• ื‘ื“ื™ื•ืง ืžืชื™ ื•ื›ื™ืฆื“ ื–ื” ืืคืฉืจื™.
07:37
In this instance, this knowledge allows us to design
129
457981
2778
ื‘ืžืงืจื” ื–ื”, ื™ื“ืข ื›ื–ื” ืžืืคืฉืจ ืœื ื• ืœืชื›ื ืŸ
07:40
novel machine architectures
130
460783
2473
ื›ืœื™-ื˜ื™ืก ื‘ืขืœื™ ืฆื•ืจื•ืช ื—ื“ืฉื ื™ื•ืช
07:43
or to design clever algorithms that gracefully handle damage,
131
463280
3668
ืื• ืœื‘ื ื•ืช ืืœื’ื•ืจื™ืชืžื™ื ื—ื›ืžื™ื ื”ืžืชืžื•ื“ื“ื™ื ื‘ื™ืขื™ืœื•ืช ืขื ื ื–ืง,
07:46
just like human athletes do,
132
466972
1700
ืžืžืฉ ื›ืžื• ืกืคื•ืจื˜ืื™ื ืื ื•ืฉื™ื™ื,
07:48
instead of building machines with redundancy.
133
468696
2264
ื‘ืžืงื•ื ืœื‘ื ื•ืช ื›ืœื™-ื˜ื™ืก ืขื ืžืขืจื›ื•ืช ื›ืคื•ืœื•ืช ืœื’ื™ื‘ื•ื™.
07:52
We can't help but hold our breath
134
472688
1657
ืื™ืŸ ืื ื• ื™ื›ื•ืœื™ื ืฉืœื ืœืขืฆื•ืจ ืืช ื ืฉื™ืžื•ืชื™ื ื•
07:54
when we watch a diver somersaulting into the water,
135
474369
2675
ื›ืืฉืจ ืื ื• ืฆื•ืคื™ื ื‘ืงื•ืคืฅ ืœื‘ืจื™ื›ื” ืขื•ืฉื” ืกืœื˜ื•ืช ื‘ืื•ื™ืจ,
07:57
or when a vaulter is twisting in the air,
136
477068
1953
ืื• ื›ืืฉืจ ืงื•ืคืฅ ื‘ืžื•ื˜ ืžืกืชื—ืจืจ ื‘ืื•ื™ืจ,
07:59
the ground fast approaching.
137
479045
1507
ื‘ื”ืชืงืจื‘ื ื‘ืžื”ื™ืจื•ืช ืœืงืจืงืข.
08:00
Will the diver be able to pull off a rip entry?
138
480576
2544
ื”ืื ื”ืงื•ืคืฅ ืœื‘ืจื™ื›ื” ื™ื™ืžื ืข ืžื›ื ื™ืกื” ื—ื–ื™ืชื™ืช ืœืžื™ื?
08:03
Will the vaulter stick the landing?
139
483144
1807
ื”ืื ื”ืงื•ืคืฅ ื‘ืžื•ื˜ ื™ื ื—ืช ื‘ืฉืœื•ื?
08:05
Suppose we want this quad here to perform a triple flip
140
485292
3180
ื ื ื™ื— ืฉื‘ืจืฆื•ื ื ื• ืฉื”ื›ืœื™ ื”ื–ื” ื™ื‘ืฆืข ื’ื™ืœื’ื•ืœ ืžืฉื•ืœืฉ
08:08
and finish off at the exact same spot that it started.
141
488496
2826
ื•ื™ืกื™ื™ื ื‘ืื•ืชื” ื ืงื•ื“ื” ืฉื‘ื” ื”ืชื—ื™ืœ.
08:11
This maneuver is going to happen so quickly
142
491736
2119
ืชื™ืžืจื•ืŸ ื–ื” ื™ืชื‘ืฆืข ื›ื” ืžื”ืจ
08:13
that we can't use position feedback to correct the motion during execution.
143
493879
3845
ืฉืื™ืŸ ืืคืฉืจื•ืช ืœื”ืฉืชืžืฉ ื‘ืžืฉื•ื‘ ืœื’ื‘ื™ ืžื™ืงื•ื ื›ื“ื™ ืœื‘ืฆืข ืชื™ืงื•ื ื™ ืชื ื•ืขื” ื‘ืžื”ืœืš ื”ื˜ื™ืกื”.
08:17
There simply isn't enough time.
144
497748
1597
ืคืฉื•ื˜ ืื™ืŸ ืžืกืคื™ืง ื–ืžืŸ.
08:19
Instead, what the quad can do is perform the maneuver blindly,
145
499782
3670
ื‘ืžืงื•ื ื–ื”, ื”ื›ืœื™ ื™ื›ื•ืœ ืœื‘ืฆืข ืืช ื”ืชื™ืžืจื•ืŸ ืขืœ ืขื™ื•ื•ืจ,
08:23
observe how it finishes the maneuver,
146
503476
2208
ืœื”ื‘ื™ืŸ ื›ื™ืฆื“ ื”ืชื™ืžืจื•ืŸ ืžืกืชื™ื™ื,
08:25
and then use that information to modify its behavior
147
505708
2477
ื•ืื– ืœื”ืฉืชืžืฉ ื‘ืžื™ื“ืข ื›ื“ื™ ืœืชืงืŸ ืืช ื”ืชื ื”ื’ื•ืชื•
08:28
so that the next flip is better.
148
508209
1638
ื›ืš ืฉื”ื’ื™ืœื’ื•ืœ ื”ื‘ื ื™ืฉืชืคืจ.
08:30
Similar to the diver and the vaulter,
149
510421
1788
ื‘ื“ื•ืžื” ืœืงื•ืคืฅ ืœื‘ืจื™ื›ื” ืื• ื”ืงื•ืคืฅ ื‘ืžื•ื˜,
08:32
it is only through repeated practice
150
512233
1895
ืจืง ื‘ืืžืฆืขื•ืช ืชื™ืจื’ื•ืœื™ื ื—ื•ื–ืจื™ื
08:34
that the maneuver can be learned and executed
151
514152
2143
ื ื™ืชืŸ ืœืœืžื•ื“ ืืช ื”ืชื™ืžืจื•ืŸ
08:36
to the highest standard.
152
516319
1242
ื•ืœื‘ืฆืขื• ื‘ืจืžื” ื”ื’ื‘ื•ื”ื” ื‘ื™ื•ืชืจ.
08:45
(Laughter)
153
525100
1269
(ืฆื—ื•ืง)
08:46
(Applause)
154
526393
3785
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
08:51
Striking a moving ball is a necessary skill in many sports.
155
531164
3448
ื”ื™ื›ื•ืœืช ืœื—ื‘ื•ื˜ ื‘ื›ื“ื•ืจ ืฉื ืข ื—ื™ื•ื ื™ืช ื‘ืขื ืคื™ ืกืคื•ืจื˜ ืจื‘ื™ื.
08:55
How do we make a machine do
156
535127
1492
ื›ื™ืฆื“ ื ื™ืชืŸ ืœื’ืจื•ื ืœืžื›ื•ื ื”
08:56
what an athlete does seemingly without effort?
157
536643
2588
ืœืขืฉื•ืช ืืช ืžื” ืฉื”ืกืคื•ืจื˜ืื™ื ืขื•ืฉื™ื ืœื›ืื•ืจื” ืœืœื ืžืืžืฅ?
09:09
(Laughter)
158
549547
1507
(ืฆื—ื•ืง)
09:16
(Applause)
159
556143
1857
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
09:20
(Applause ends)
160
560500
1298
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื ืžืกืชื™ื™ืžื•ืช)
09:22
This quad has a racket strapped onto its head
161
562542
2626
ืœืžืกื•ืง ื–ื” ื™ืฉ ืžื—ื‘ื˜ ื”ืžื•ืฆืžื“ ืœืจืืฉื•,
09:25
with a sweet spot roughly the size of an apple, so not too large.
162
565192
3463
ื‘ืขืœ ืžืฉื˜ื— ื”ื—ื–ืจื” ื‘ืขืจืš ื‘ื’ื•ื“ืœ ืชืคื•ื—, ืฉื–ื” ืœื ื’ื“ื•ืœ ื‘ืžื™ื•ื—ื“.
09:29
The following calculations are made every 20 milliseconds,
163
569017
2805
ื”ื—ื™ืฉื•ื‘ื™ื ื”ื‘ืื™ื ื ืขืจื›ื™ื ื›ืœ 20 ืžื™ืœื™ืฉื ื™ื•ืช,
09:31
or 50 times per second.
164
571846
1317
ืื• 50 ืคืขื ื‘ืฉื ื™ื”.
09:33
We first figure out where the ball is going.
165
573602
2275
ืชื—ื™ืœื” ืื ื• ืžื—ืฉื‘ื™ื ืœืืŸ ื”ื›ื“ื•ืจ ื ืข.
09:36
We then next calculate how the quad should hit the ball
166
576414
2621
ืœืื—ืจ-ืžื›ืŸ ืžื—ืฉื‘ื™ื ื›ื™ืฆื“ ื”ืžืกื•ืง ืฆืจื™ืš ืœืคื’ื•ืข ื‘ื›ื“ื•ืจ
09:39
so that it flies to where it was thrown from.
167
579059
2485
ื›ืš ืฉื™ืขื•ืฃ ื‘ื—ื–ืจื” ืœืื•ืชื• ืžืงื•ื ืฉืžืžื ื• ื™ืฆื.
09:41
Third, a trajectory is planned that carries the quad
168
581950
4432
ืฉืœื™ืฉื™ืช, ืžื—ื•ืฉื‘ ื”ืžืกืœื•ืœ ืฉืžื•ื‘ื™ืœ ืืช ื”ืžืกื•ืง ืžื”ืžื™ืงื•ื
09:46
from its current state to the impact point with the ball.
169
586406
3000
ื”ื ื•ื›ื—ื™ ืฉืœื• ืœื ืงื•ื“ืช ื”ืžืคื’ืฉ ืขื ื”ื›ื“ื•ืจ.
09:49
Fourth, we only execute 20 milliseconds' worth of that strategy.
170
589430
3579
ืจื‘ื™ืขื™ืช, ืื ื• ืžื•ืฆื™ืื™ื ืœืคื•ืขืœ ืจืง 20 ืžื™ืœื™ืฉื ื™ื•ืช ืžืชื•ืš ืื•ืชื• ื—ื™ืฉื•ื‘.
09:53
Twenty milliseconds later, the whole process is repeated
171
593366
2793
ื‘ืขื•ื“ 20 ืžื™ืœื™ืฉื ื™ื•ืช, ื—ื•ื–ืจื™ื ืขืœ ื›ืœ ื”ืชื”ืœื™ืš
09:56
until the quad strikes the ball.
172
596183
1767
ืขื“ ืฉื”ืžืกื•ืง ืžื›ื” ืืช ื”ื›ื“ื•ืจ ื‘ื—ื–ืจื”.
10:07
(Applause)
173
607290
2404
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
10:10
Machines can not only perform dynamic maneuvers on their own,
174
610131
3423
ื”ื›ืœื™ื ืœื ืจืง ืžื‘ืฆืขื™ื ืชื™ืžืจื•ื ื™ื ื“ื™ื ืžื™ื™ื ื‘ืื•ืคืŸ ืขืฆืžืื™,
10:13
they can do it collectively.
175
613578
1839
ื”ื ืืฃ ื™ื›ื•ืœื™ื ืœืขืฉื•ืช ื–ืืช ื‘ืฆื•ื•ืชื.
10:15
These three quads are cooperatively carrying a sky net.
176
615441
3587
ืฉืœื•ืฉืช ื”ืžืกื•ืงื™ื ื”ืœืœื• ื ื•ืฉืื™ื ื‘ืฆื•ื•ืชื ืจืฉืช.
10:27
(Applause)
177
627589
1679
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
10:32
(Applause ends)
178
632372
1589
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื ืžืกืชื™ื™ืžื•ืช)
10:33
They perform an extremely dynamic and collective maneuver
179
633985
4341
ื”ื ืžื‘ืฆืขื™ื ื‘ืฆื•ื•ืชื ืชื™ืžืจื•ืŸ ื—ื“ ื‘ืื•ืคืŸ ืงื™ืฆื•ื ื™
10:38
to launch the ball back to me.
180
638350
1668
ื›ื“ื™ ืœืฉื’ืจ ืืช ื”ื›ื“ื•ืจ ื‘ื—ื–ืจื” ืืœื™ื™.
10:40
Notice that, at full extension, these quads are vertical.
181
640042
3146
ืฉื™ืžื• ืœื‘ ืฉื‘ืžืชื™ื—ื” ื”ืžื›ืกื™ืžืœื™ืช, ื”ืžืกื•ืงื™ื ื‘ืžืฆื‘ ืื ื›ื™.
10:48
(Applause)
182
648486
1748
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
10:50
In fact, when fully extended,
183
650258
2143
ื‘ืขืฆื, ื‘ืžืชื™ื—ื” ื”ืžื›ืกื™ืžืœื™ืช,
10:52
this is roughly five times greater than what a bungee jumper feels
184
652425
3263
ื–ื” ื‘ืขืจืš ืคื™-5 ื™ื•ืชืจ ื—ื–ืง ืžืžื” ืฉื—ืฉ ืงื•ืคืฅ ื‘ืื ื’'ื™
10:55
at the end of their launch.
185
655712
1970
ื‘ืกื•ืฃ ืงืคื™ืฆืชื•.
11:03
The algorithms to do this are very similar
186
663203
2459
ื”ืืœื’ื•ืจื™ืชืžื™ื ืฉืžื‘ืฆืขื™ื ื–ืืช
11:05
to what the single quad used to hit the ball back to me.
187
665686
2915
ื“ื•ืžื™ื ืžืื•ื“ ืœืืœื• ืฉืฉื™ืžืฉื• ืœื”ื›ื•ืช ื‘ื›ื“ื•ืจ ื‘ื—ื–ืจื” ืืœื™ื™.
11:09
Mathematical models are used to continuously re-plan
188
669070
2840
ืžื•ื“ืœื™ื ืžืชืžื˜ื™ื™ื ืžืฉืžืฉื™ื ืœืชื™ื›ื ื•ืŸ
11:11
a cooperative strategy 50 times per second.
189
671934
2831
ืžื—ื–ื•ืจื™ ื•ืžืชืžืฉืš ืฉืœ ืืกื˜ืจื˜ื’ื™ื™ื” ืžืฉื•ืชืคืช 50 ืคืขื ื‘ืฉื ื™ื”.
11:16
Everything we have seen so far has been about the machines and their capabilities.
190
676506
4259
ื›ืœ ืžื” ืฉืจืื™ื ื• ืขื“ ืขืชื” ื”ื™ื” ืงืฉื•ืจ ื‘ืžื›ื•ื ื•ืช ื•ื‘ื™ื›ื•ืœื•ืชื™ื”ืŸ.
11:20
What happens when we couple this machine athleticism
191
680789
2992
ืžื” ืงื•ืจื” ื›ืืฉืจ ืžื–ื•ื•ื’ื™ื ืืช ื”ื™ื›ื•ืœืช ื”ืกืคื•ืจื˜ื™ื‘ื™ืช ืฉืœ ืžื›ื•ื ื” ื–ื•
11:23
with that of a human being?
192
683805
1483
ืขื ื–ื• ืฉืœ ื”ืื“ื?
11:25
What I have in front of me is a commercial gesture sensor
193
685765
3596
ืžื” ืฉืžื•ื ื— ืœืคื ื™ื™ ื–ื” ื—ื™ื™ืฉืŸ ืžืกื—ืจื™ ืฉืœ ืชื ื•ืขื•ืช ืื ื•ืฉื™ื•ืช
11:29
mainly used in gaming.
194
689385
1285
ืืฉืจ ืžืฉืžืฉ ื‘ืขื™ืงืจ ื‘ืžืฉื—ืงื™ื.
11:30
It can recognize what my various body parts
195
690694
2048
ื”ื•ื ืžืกื•ื’ืœ ืœื–ื”ื•ืช ื‘ื–ืžืŸ-ืืžืช ืžื” ื”ืื™ื‘ืจื™ื ื”ืฉื•ื ื™ื ืฉืœ ื’ื•ืคื™
11:32
are doing in real time.
196
692766
1317
ืขื•ืฉื™ื ื‘ื–ืžืŸ ืืžืช.
11:34
Similar to the pointer that I used earlier,
197
694710
2079
ื‘ื“ื•ืžื” ืœืžืฆื‘ื™ืข ื‘ื• ื”ืฉืชืžืฉืชื™ ืžืงื•ื“ื,
11:36
we can use this as inputs to the system.
198
696813
2222
ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ื–ื” ื›ืงืœื˜ ืฉืœ ื”ืžืขืจื›ืช.
11:39
We now have a natural way of interacting
199
699454
2372
ื›ืขืช ื™ืฉ ื‘ื™ื“ื™ื ื• ื“ืจืš ื˜ื‘ืขื™ืช ืœืชืงืฉืจ
11:41
with the raw athleticism of these quads with my gestures.
200
701850
3261
ืขื ื”ื™ื›ื•ืœืช ื”ืืชืœื˜ื™ืช ื”ื’ื•ืœืžื™ืช ืฉืœ ื”ืžืกื•ืงื™ื ื”ืœืœื• ืœื‘ื™ืŸ ืคืขื•ืœื•ืช ื”ื’ื•ืฃ ืฉืœื™.
12:22
(Applause)
201
742715
3857
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
12:36
Interaction doesn't have to be virtual.
202
756354
2016
ื”ืžื’ืข ืื™ื ื• ื—ื™ื™ื‘ ืœื”ื™ื•ืช ื•ื™ืจื˜ื•ืืœื™.
12:38
It can be physical.
203
758394
1190
ื”ื•ื ื™ื›ื•ืœ ืœื”ื™ื•ืช ืคื™ื–ื™.
12:39
Take this quad, for example.
204
759989
1515
ื ื™ืงื— ืœื“ื•ื’ืžื ืžืกื•ืง ื–ื”.
12:41
It's trying to stay at a fixed point in space.
205
761893
2223
ื”ื•ื ืžื ืกื” ืœื”ื™ืฉืืจ ื‘ื ืงื•ื“ื” ืงื‘ื•ืขื” ื‘ื—ืœืœ.
12:44
If I try to move it out of the way, it fights me,
206
764933
3689
ืื ืื ื™ ืžื ืกื” ืœื”ืกื™ื˜ื• ืžืžืงื•ืžื•, ื”ื•ื ืžืชื ื’ื“
12:48
and moves back to where it wants to be.
207
768646
1874
ื•ื ืข ื‘ื—ื–ืจื” ืœื”ื™ื›ืŸ ืฉื”ื•ื ืจื•ืฆื” ืœื”ื™ื•ืช.
12:52
We can change this behavior, however.
208
772254
2087
ืื‘ืœ ื ื™ืชืŸ ืœืฉื ื•ืช ื”ืชื ื”ื’ื•ืช ื–ื•.
12:55
We can use mathematical models
209
775373
1675
ืื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ืžื•ื“ืœื™ื ืžืชืžื˜ื™ื™ื
12:57
to estimate the force that I'm applying to the quad.
210
777072
2664
ื›ื“ื™ ืœืืžื•ื“ ืืช ื”ื›ื•ื— ืฉืื ื™ ืžืคืขื™ืœ ืขืœ ื”ืžืกื•ืง.
13:00
Once we know this force, we can also change the laws of physics,
211
780205
3214
ื‘ืจื’ืข ืฉื™ื•ื“ืขื™ื ืืช ื”ื›ื•ื—, ื ื™ืชืŸ ื’ื ืœืฉื ื•ืช ืืช
13:03
as far as the quad is concerned, of course.
212
783443
2087
ื—ื•ืงื™ ื”ืคื™ื–ื™ืงื”, ื›ื›ืœ ืฉื–ื” ื ื•ื’ืข ืœืžืกื•ืง ื›ืžื•ื‘ืŸ.
13:07
Here, the quad is behaving as if it were in a viscous fluid.
213
787942
3309
ื›ืืŸ ื”ืžืกื•ืง ืžืชื ื”ื’ ื›ืื™ืœื• ื”ื•ื ื ืžืฆื ื‘ื ื•ื–ืœ ืฆืžื™ื’.
13:14
We now have an intimate way of interacting with a machine.
214
794570
3719
ื›ืขืช ื™ืฉ ืœื ื• ื™ื›ื•ืœืช ืœื™ืฆื™ืจืช ืžื’ืข ื”ื“ื•ืง ืขื ื”ืžื›ื•ื ื”.
13:18
I will use this new capability to position
215
798593
2482
ืืฉืชืžืฉ ื‘ื™ื›ื•ืœืช ื—ื“ืฉื” ื–ื• ื›ื“ื™ ืœืžืงื
13:21
this camera-carrying quad to the appropriate location
216
801099
2553
ืืช ื”ืžืกื•ืง ื”ื–ื” ืฉื ื•ืฉื ืžืฆืœืžื” ื‘ืžื™ืงื•ื ืžืชืื™ื
13:23
for filming the remainder of this demonstration.
217
803676
2428
ื›ื“ื™ ืœืฆืœื ืืช ืžื” ืฉื ื•ืชืจ ืžื”ืชืฆื•ื’ื”.
13:36
So we can physically interact with these quads
218
816583
2383
ืื ื• ืžืกื•ื’ืœื™ื ืœื‘ื•ื ื‘ืžื’ืข ืคื™ื–ื™ ืขื ื”ืžืกื•ืงื™ื
13:38
and we can change the laws of physics.
219
818990
1960
ื•ืื ื• ืžืกื•ื’ืœื™ื ืœืฉื ื•ืช ืืช ื—ื•ืงื™ ื”ืคื™ื–ื™ืงื”.
13:41
Let's have a little bit of fun with this.
220
821393
1977
ื”ื‘ื” ื ื”ื ื” ืžื–ื” ืงืฆืช.
13:43
For what you will see next,
221
823539
1369
ืžื” ืฉืชืจืื• ืขื›ืฉื™ื•, ืžืกื•ืงื™ื ื”ืœืœื•
13:44
these quads will initially behave as if they were on Pluto.
222
824932
3294
ื™ืชื ื”ื’ื• ืชื—ื™ืœื” ื›ืื™ืœื• ื”ื ื ืžืฆืื™ื ืขืœ ืคืœื•ื˜ื•.
13:48
As time goes on, gravity will be increased
223
828906
2288
ืขื ื—ืœื•ืฃ ื”ื–ืžืŸ, ื›ื•ื—-ื”ืžืฉื™ื›ื” ื™ื•ื’ื‘ืจ
13:51
until we're all back on planet Earth,
224
831218
2025
ืขื“ ืฉื”ื ื™ื”ื™ื• ื‘ื—ื–ืจื” ืขืœ ื›ื“ื•ืจ-ื”ืืจืฅ,
13:53
but I assure you we won't get there.
225
833267
1766
ืื‘ืœ ืื ื™ ืžื‘ื˜ื™ื—ื›ื ืฉืœื ื ื’ื™ืข ืœืฉื.
13:55
Okay, here goes.
226
835462
1024
ื˜ื•ื‘, ื”ื ื” ื–ื” ืžืชื—ื™ืœ.
14:04
(Laughter)
227
844897
1551
(ืฆื—ื•ืง)
14:34
(Laughter)
228
874778
2960
(ืฆื—ื•ืง)
14:37
(Applause)
229
877858
3761
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
14:41
Whew!
230
881760
1066
ื™ื•!
14:47
You're all thinking now,
231
887045
1227
ื›ื•ืœื›ื ื•ื“ืื™ ื—ื•ืฉื‘ื™ื ืขื›ืฉื™ื•,
14:48
these guys are having way too much fun,
232
888296
1951
ืื ืฉื™ื ืืœื” ื ื”ื ื™ื ื™ื•ืชืจ ืžื“ื™,
14:50
and you're probably also asking yourself,
233
890271
2055
ื•ืงืจื•ื‘ ืœื•ื“ืื™ ืืชื ืฉื•ืืœื™ื ืืช ืขืฆืžื›ื,
14:52
why exactly are they building machine athletes?
234
892350
2801
ืžื“ื•ืข ื”ื ื‘ื›ืœืœ ื‘ื•ื ื™ื ืืชืœื˜ื™ื ืžื›ื ื™ื™ื?
14:56
Some conjecture that the role of play in the animal kingdom
235
896286
2810
ื™ืฉ ื”ืื•ืžืจื™ื ืฉืชืคืงื™ื“ ื”ืžืฉื—ืง ื‘ืžืžืœื›ืช ื”ื—ื™ื•ืช
14:59
is to hone skills and develop capabilities.
236
899120
2137
ื”ื•ื ืœื”ืฉื—ื™ื– ื›ื™ืฉื•ืจื™ื ื•ืœืคืชื— ื™ื›ื•ืœื•ืช.
15:01
Others think that it has more of a social role,
237
901690
2215
ืื—ืจื™ื ืกื‘ื•ืจื™ื ืฉื™ืฉ ืœื• ื™ื•ืชืจ ืชืคืงื™ื“ ื—ื‘ืจืชื™,
15:03
that it's used to bind the group.
238
903929
1650
ืฉื”ื•ื ืžืฉืžืฉ ืœื’ื™ื‘ื•ืฉ ื”ืงื‘ื•ืฆื”.
15:05
Similarly, we use the analogy of sports and athleticism
239
905603
3336
ื‘ืื•ืคืŸ ื“ื•ืžื”, ืื ื• ืžืฉืชืžืฉื™ื ื‘ืื ืœื•ื’ื™ื” ืฉืœ ืกืคื•ืจื˜ ื•ืืชืœื˜ื™ื•ืช
15:08
to create new algorithms for machines
240
908963
2055
ื›ื“ื™ ืœื™ืฆื•ืจ ืืœื’ื•ืจื™ืชืžื™ื ื—ื“ืฉื™ื ืœืžื›ื•ื ื•ืช
15:11
to push them to their limits.
241
911042
1443
ื•ืœื”ื‘ื™ืืŸ ืœืงืฆื” ื’ื‘ื•ืœ ื™ื›ื•ืœืชืŸ.
15:13
What impact will the speed of machines have on our way of life?
242
913350
3048
ืื™ื–ื• ื”ืฉืคืขื” ืชื”ื™ื” ืœืžื”ื™ืจื•ืช ืฉืœ ื”ืžื›ื•ื ื•ืช ืขืœ ื“ืจืš ื—ื™ื™ื ื•?
15:16
Like all our past creations and innovations,
243
916866
2365
ื›ืžื• ื›ืœ ื™ืฆื™ืจื•ืชื™ื ื• ื•ื”ืžืฆืื•ืชื™ื ื• ื‘ืขื‘ืจ,
15:19
they may be used to improve the human condition
244
919255
2803
ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ื”ืŸ ืœืฉื™ืคื•ืจ ื”ืชื ืื™ื ื”ืื ื•ืฉื™ื™ื
15:22
or they may be misused and abused.
245
922082
1974
ืื• ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ื”ืŸ ืœืจืขื”.
15:24
This is not a technical choice we are faced with;
246
924595
2302
ืื™ืŸ ื–ื• ื‘ื—ื™ืจื” ื˜ื›ื ื•ืœื•ื’ื™ืช ืฉืื ื• ืขื•ืžื“ื™ื ื‘ืคื ื™ื”;
15:26
it's a social one.
247
926921
1015
ื–ื•ื”ื™ ื‘ื—ื™ืจื” ื—ื‘ืจืชื™ืช-ืžื•ืกืจื™ืช.
15:28
Let's make the right choice,
248
928332
1351
ื”ื‘ื” ื ืขืฉื” ืืช ื”ื‘ื—ื™ืจื” ื”ื ื›ื•ื ื”,
15:29
the choice that brings out the best in the future of machines,
249
929707
2912
ื”ื‘ื—ื™ืจื” ืฉืžื•ืฆื™ืื” ืืช ื”ืžื™ื˜ื‘ ืžืขืชื™ื“ืŸ ืฉืœ ื”ืžื›ื•ื ื•ืช ื”ืœืœื•,
15:32
just like athleticism in sports can bring out the best in us.
250
932643
2865
ื‘ื“ื™ื•ืง ื›ืžื• ืฉืืชืœื˜ื™ื•ืช ื‘ืกืคื•ืจื˜ ื™ื›ื•ืœื” ืœื”ื•ืฆื™ื ืืช ื”ืžื™ื˜ื‘ ืžืชื•ื›ื ื•.
15:36
Let me introduce you to the wizards behind the green curtain.
251
936380
3075
ืืฆื™ื’ ื‘ืคื ื™ื›ื ืืช ืืฉืคื™ ื”ืžื—ืฉื‘ ืฉืžืื—ื•ืจื™ ื”ืžืกืš ื”ื™ืจื•ืง.
15:39
They're the current members of the Flying Machine Arena research team.
252
939479
3334
ื”ื ื”ื—ื‘ืจื™ื ื”ื ื•ื›ื—ื™ื™ื ื‘ืงื‘ื•ืฆืช ื”ืžื—ืงืจ "ื–ื™ืจืช ื”ืžื›ื•ื ื•ืช ื”ืžืขื•ืคืคื•ืช".
15:42
(Applause)
253
942837
4217
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
15:47
Federico Augugliaro, Dario Brescianini,
254
947267
2278
ืคื“ืจื™ืงื• ืื™ื•ื’ื•ื’ืœื™ืืจื™ื•, ื“ืืจื™ื• ื‘ืจืกื™ืื ื™ื ื™, ืžืจื›ื•ืก ื”ืŸ,
15:49
Markus Hehn, Sergei Lupashin, Mark Muller and Robin Ritz.
255
949569
3425
ืกืจื’'ื™ื™ ืœื•ืคืืฉื™ืŸ, ืžืืจืง ืžื•ืœืจ ื•ืจื•ื‘ื™ืŸ ืจื™ื˜ืก.
15:53
Look out for them. They're destined for great things.
256
953018
2524
ื”ืกืชื›ืœื• ืขืœื™ื”ื. ื”ื ื ื•ืขื“ื• ืœื’ื“ื•ืœื•ืช.
15:55
Thank you.
257
955566
1054
ืชื•ื“ื” ืœื›ื.
15:56
(Applause)
258
956644
3456
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7