TEDWomen 2015
Robin Murphy: These robots come to the rescue after a disaster
Filmed:
Readability: 4.3
1,125,212 views
When disaster strikes, who's first on the scene? More and more, it’s a robot. In her lab, Robin Murphy builds robots that fly, tunnel, swim and crawl through disaster scenes, helping firefighters and rescue workers save more lives safely -- and help communities return to normal up to three years faster.
Robin Murphy - Disaster roboticist
Robin Murphy researches robots -- ground, aerial and marine -- that can help out during disasters. Full bio
Robin Murphy researches robots -- ground, aerial and marine -- that can help out during disasters. Full bio
Double-click the English transcript below to play the video.
00:12
Over a million people are killed
each year in disasters.
each year in disasters.
0
999
4976
00:17
Two and a half million people
will be permanently disabled or displaced,
will be permanently disabled or displaced,
1
5999
5234
00:23
and the communities will take
20 to 30 years to recover
20 to 30 years to recover
2
11257
3924
00:27
and billions of economic losses.
3
15205
2401
00:31
If you can reduce
the initial response by one day,
the initial response by one day,
4
19054
4371
00:35
you can reduce the overall recovery
5
23449
3527
00:39
by a thousand days, or three years.
6
27000
2808
00:41
See how that works?
7
29832
1753
00:43
If the initial responders
can get in, save lives,
can get in, save lives,
8
31609
2850
00:46
mitigate whatever flooding
danger there is,
danger there is,
9
34483
2516
00:49
that means the other groups can get in
10
37023
2284
00:51
to restore the water,
the roads, the electricity,
the roads, the electricity,
11
39331
2985
00:54
which means then the construction people,
the insurance agents,
the insurance agents,
12
42340
3116
00:57
all of them can get in
to rebuild the houses,
to rebuild the houses,
13
45480
2614
01:00
which then means
you can restore the economy,
you can restore the economy,
14
48118
2988
01:03
and maybe even make it better
and more resilient to the next disaster.
and more resilient to the next disaster.
15
51130
4873
01:09
A major insurance company told me
16
57899
1796
01:11
that if they can get a homeowner's claim
processed one day earlier,
processed one day earlier,
17
59719
5112
01:16
it'll make a difference of six months
18
64855
2120
01:18
in that person getting
their home repaired.
their home repaired.
19
66999
2240
01:22
And that's why I do disaster robotics --
20
70118
2329
01:24
because robots can
make a disaster go away faster.
make a disaster go away faster.
21
72471
4499
01:30
Now, you've already seen
a couple of these.
a couple of these.
22
78305
2017
01:32
These are the UAVs.
23
80346
1720
01:34
These are two types of UAVs:
24
82090
1861
01:35
a rotorcraft, or hummingbird;
25
83975
1865
01:37
a fixed-wing, a hawk.
26
85864
2070
01:39
And they're used extensively since 2005 --
27
87958
3390
01:43
Hurricane Katrina.
28
91372
1175
01:44
Let me show you how this hummingbird,
this rotorcraft, works.
this rotorcraft, works.
29
92571
3324
01:47
Fantastic for structural engineers.
30
95919
3056
01:50
Being able to see damage from angles you
can't get from binoculars on the ground
can't get from binoculars on the ground
31
98999
4420
01:55
or from a satellite image,
32
103443
1310
01:56
or anything flying at a higher angle.
33
104777
3192
02:00
But it's not just structural engineers
and insurance people who need this.
and insurance people who need this.
34
108530
4372
02:04
You've got things
like this fixed-wing, this hawk.
like this fixed-wing, this hawk.
35
112926
2467
02:07
Now, this hawk can be used
for geospatial surveys.
for geospatial surveys.
36
115417
3271
02:10
That's where you're
pulling imagery together
pulling imagery together
37
118712
2849
02:13
and getting 3D reconstruction.
38
121585
1825
02:15
We used both of these at the Oso mudslides
up in Washington State,
up in Washington State,
39
123434
4074
02:19
because the big problem
40
127532
2199
02:21
was geospatial and hydrological
understanding of the disaster --
understanding of the disaster --
41
129755
3158
02:24
not the search and rescue.
42
132937
1246
02:26
The search and rescue teams
had it under control
had it under control
43
134207
2319
02:28
and knew what they were doing.
44
136550
1478
02:30
The bigger problem was that river
and mudslide might wipe them out
and mudslide might wipe them out
45
138052
3870
02:33
and flood the responders.
46
141946
1494
02:35
And not only was it challenging
to the responders and property damage,
to the responders and property damage,
47
143464
3742
02:39
it's also putting at risk
the future of salmon fishing
the future of salmon fishing
48
147230
3279
02:42
along that part of Washington State.
49
150533
1716
02:44
So they needed to understand
what was going on.
what was going on.
50
152273
2476
02:46
In seven hours, going from Arlington,
51
154773
2609
02:49
driving from the Incident Command Post
to the site, flying the UAVs,
to the site, flying the UAVs,
52
157406
4814
02:54
processing the data, driving back
to Arlington command post --
to Arlington command post --
53
162244
3731
02:57
seven hours.
54
165999
1406
02:59
We gave them in seven hours
data that they could take
data that they could take
55
167429
3534
03:02
only two to three days
to get any other way --
to get any other way --
56
170987
3750
03:06
and at higher resolution.
57
174761
1837
03:09
It's a game changer.
58
177064
1373
03:11
And don't just think about the UAVs.
59
179773
2069
03:13
I mean, they are sexy -- but remember,
60
181866
2628
03:16
80 percent of the world's
population lives by water,
population lives by water,
61
184518
3336
03:19
and that means our critical
infrastructure is underwater --
infrastructure is underwater --
62
187878
2864
03:22
the parts that we can't get to,
like the bridges and things like that.
like the bridges and things like that.
63
190766
3286
03:26
And that's why we have
unmanned marine vehicles,
unmanned marine vehicles,
64
194076
2592
03:28
one type of which you've already met,
which is SARbot, a square dolphin.
which is SARbot, a square dolphin.
65
196692
4452
03:33
It goes underwater and uses sonar.
66
201168
2696
03:35
Well, why are marine vehicles so important
67
203888
2333
03:38
and why are they very, very important?
68
206245
3124
03:41
They get overlooked.
69
209393
1516
03:42
Think about the Japanese tsunami --
70
210933
2402
03:45
400 miles of coastland totally devastated,
71
213359
3966
03:49
twice the amount of coastland devastated
by Hurricane Katrina in the United States.
by Hurricane Katrina in the United States.
72
217349
4696
03:54
You're talking about your bridges,
your pipelines, your ports -- wiped out.
your pipelines, your ports -- wiped out.
73
222069
3856
03:57
And if you don't have a port,
74
225949
1612
03:59
you don't have a way
to get in enough relief supplies
to get in enough relief supplies
75
227585
3246
04:02
to support a population.
76
230855
1221
04:04
That was a huge problem
at the Haiti earthquake.
at the Haiti earthquake.
77
232100
2572
04:07
So we need marine vehicles.
78
235565
2122
04:09
Now, let's look at a viewpoint
from the SARbot
from the SARbot
79
237711
2431
04:12
of what they were seeing.
80
240166
1471
04:13
We were working on a fishing port.
81
241661
2246
04:15
We were able to reopen that fishing port,
using her sonar, in four hours.
using her sonar, in four hours.
82
243931
5716
04:21
That fishing port was told
it was going to be six months
it was going to be six months
83
249671
2653
04:24
before they could get
a manual team of divers in,
a manual team of divers in,
84
252348
2706
04:27
and it was going to take
the divers two weeks.
the divers two weeks.
85
255078
2687
04:29
They were going to miss
the fall fishing season,
the fall fishing season,
86
257789
2294
04:32
which was the major economy for that part,
which is kind of like their Cape Cod.
which is kind of like their Cape Cod.
87
260107
4226
04:36
UMVs, very important.
88
264357
2268
04:38
But you know, all the robots
I've shown you have been small,
I've shown you have been small,
89
266649
3248
04:41
and that's because robots
don't do things that people do.
don't do things that people do.
90
269921
3777
04:45
They go places people can't go.
91
273722
2253
04:47
And a great example of that is Bujold.
92
275999
2976
04:50
Unmanned ground vehicles
are particularly small,
are particularly small,
93
278999
2976
04:53
so Bujold --
94
281999
1318
04:55
(Laughter)
95
283341
1508
04:56
Say hello to Bujold.
96
284873
1183
04:58
(Laughter)
97
286080
2684
05:01
Bujold was used extensively
at the World Trade Center
at the World Trade Center
98
289606
3369
05:04
to go through Towers 1, 2 and 4.
99
292999
2380
05:07
You're climbing into the rubble,
rappelling down, going deep in spaces.
rappelling down, going deep in spaces.
100
295403
4645
05:12
And just to see the World Trade Center
from Bujold's viewpoint, look at this.
from Bujold's viewpoint, look at this.
101
300072
4559
05:16
You're talking about a disaster
where you can't fit a person or a dog --
where you can't fit a person or a dog --
102
304655
5069
05:21
and it's on fire.
103
309748
2142
05:23
The only hope of getting
to a survivor way in the basement,
to a survivor way in the basement,
104
311914
3734
05:27
you have to go through things
that are on fire.
that are on fire.
105
315672
2195
05:29
It was so hot, on one of the robots,
the tracks began to melt and come off.
the tracks began to melt and come off.
106
317891
4504
05:35
Robots don't replace people or dogs,
107
323161
2788
05:37
or hummingbirds or hawks or dolphins.
108
325973
2521
05:40
They do things new.
109
328518
2036
05:42
They assist the responders,
the experts, in new and innovative ways.
the experts, in new and innovative ways.
110
330578
5689
05:48
The biggest problem is not
making the robots smaller, though.
making the robots smaller, though.
111
336291
4429
05:52
It's not making them more heat-resistant.
112
340744
2231
05:54
It's not making more sensors.
113
342999
1757
05:56
The biggest problem is the data,
the informatics,
the informatics,
114
344780
3610
06:00
because these people need to get
the right data at the right time.
the right data at the right time.
115
348414
4036
06:04
So wouldn't it be great if we could have
experts immediately access the robots
experts immediately access the robots
116
352474
5501
06:09
without having to waste any time
of driving to the site,
of driving to the site,
117
357999
2976
06:12
so whoever's there,
use their robots over the Internet.
use their robots over the Internet.
118
360999
2933
06:15
Well, let's think about that.
119
363956
1382
06:17
Let's think about a chemical
train derailment in a rural county.
train derailment in a rural county.
120
365362
3372
06:20
What are the odds that the experts,
your chemical engineer,
your chemical engineer,
121
368758
3842
06:24
your railroad transportation engineers,
122
372624
1862
06:26
have been trained on whatever UAV
that particular county happens to have?
that particular county happens to have?
123
374510
4861
06:31
Probably, like, none.
124
379395
1533
06:32
So we're using these kinds of interfaces
125
380952
2390
06:35
to allow people to use the robots
without knowing what robot they're using,
without knowing what robot they're using,
126
383366
4340
06:39
or even if they're using a robot or not.
127
387730
2786
06:44
What the robots give you,
what they give the experts, is data.
what they give the experts, is data.
128
392252
6034
06:50
The problem becomes:
who gets what data when?
who gets what data when?
129
398310
3580
06:53
One thing to do is to ship
all the information to everybody
all the information to everybody
130
401914
3893
06:57
and let them sort it out.
131
405831
1512
06:59
Well, the problem with that
is it overwhelms the networks,
is it overwhelms the networks,
132
407367
3856
07:03
and worse yet, it overwhelms
the cognitive abilities
the cognitive abilities
133
411247
3686
07:06
of each of the people trying to get
that one nugget of information
that one nugget of information
134
414957
4509
07:11
they need to make the decision
that's going to make the difference.
that's going to make the difference.
135
419490
3805
07:15
So we need to think
about those kinds of challenges.
about those kinds of challenges.
136
423999
3094
07:19
So it's the data.
137
427117
1265
07:20
Going back to the World Trade Center,
138
428406
2236
07:22
we tried to solve that problem
by just recording the data from Bujold
by just recording the data from Bujold
139
430666
4309
07:26
only when she was deep in the rubble,
140
434999
1976
07:28
because that's what the USAR team
said they wanted.
said they wanted.
141
436999
2564
07:32
What we didn't know at the time
142
440523
2588
07:35
was that the civil engineers
would have loved,
would have loved,
143
443135
2598
07:37
needed the data as we recorded
the box beams, the serial numbers,
the box beams, the serial numbers,
144
445757
4218
07:41
the locations, as we went into the rubble.
145
449999
2619
07:45
We lost valuable data.
146
453343
1467
07:46
So the challenge is getting all the data
147
454834
2379
07:49
and getting it to the right people.
148
457237
2180
07:51
Now, here's another reason.
149
459441
2077
07:53
We've learned that some buildings --
150
461542
2311
07:55
things like schools,
hospitals, city halls --
hospitals, city halls --
151
463877
3272
07:59
get inspected four times
by different agencies
by different agencies
152
467173
4019
08:03
throughout the response phases.
153
471216
2250
08:06
Now, we're looking, if we can get
the data from the robots to share,
the data from the robots to share,
154
474156
3269
08:09
not only can we do things like
compress that sequence of phases
compress that sequence of phases
155
477449
4850
08:14
to shorten the response time,
156
482323
1653
08:16
but now we can begin
to do the response in parallel.
to do the response in parallel.
157
484000
4358
08:20
Everybody can see the data.
158
488382
1379
08:21
We can shorten it that way.
159
489785
1462
08:23
So really, "disaster robotics"
is a misnomer.
is a misnomer.
160
491999
3406
08:28
It's not about the robots.
161
496227
1895
08:30
It's about the data.
162
498146
1882
08:32
(Applause)
163
500433
3330
08:35
So my challenge to you:
164
503787
1957
08:37
the next time you hear about a disaster,
165
505768
2300
08:40
look for the robots.
166
508092
1325
08:41
They may be underground,
they may be underwater,
they may be underwater,
167
509441
3125
08:44
they may be in the sky,
168
512590
1694
08:46
but they should be there.
169
514308
1965
08:48
Look for the robots,
170
516297
1167
08:49
because robots are coming to the rescue.
171
517488
2747
08:52
(Applause)
172
520259
5661
ABOUT THE SPEAKER
Robin Murphy - Disaster roboticistRobin Murphy researches robots -- ground, aerial and marine -- that can help out during disasters.
Why you should listen
Robin Murphy imagines how robots can do tasks no human could amid scenes of disaster hard to imagine, from the World Trade Center disaster to Hurricane Katrina to the Fukushima Daiichi nuclear emergency. In her recent book, Disaster Robotics, she lays out her research into the problem, which pulls together artificial intelligence, robotics and human-robot interaction.
At Texas A&M, Murphy is the director of the Center for Robot-Assisted Search and Rescue and the Center for Emergency Informatics. She also co-founded the IEEE Robotics and Automation Society’s Technical Committee on Safety Security and Rescue Robotics and its annual conference. Her field work, combined with technology transfer and research community-building activities, led to her receiving the 2014 ACM Eugene L. Lawler Award for Humanitarian Contributions within Computer Science and Informatics.
Robin Murphy | Speaker | TED.com