1
00:00:00,999 --> 00:00:05,975
Over a million people are killed
each year in disasters.

2
00:00:05,999 --> 00:00:11,233
Two and a half million people
will be permanently disabled or displaced,

3
00:00:11,257 --> 00:00:15,181
and the communities will take
20 to 30 years to recover

4
00:00:15,205 --> 00:00:17,606
and billions of economic losses.

5
00:00:19,054 --> 00:00:23,425
If you can reduce
the initial response by one day,

6
00:00:23,449 --> 00:00:26,976
you can reduce the overall recovery

7
00:00:27,000 --> 00:00:29,808
by a thousand days, or three years.

8
00:00:29,832 --> 00:00:31,585
See how that works?

9
00:00:31,609 --> 00:00:34,459
If the initial responders
can get in, save lives,

10
00:00:34,483 --> 00:00:36,999
mitigate whatever flooding
danger there is,

11
00:00:37,023 --> 00:00:39,307
that means the other groups can get in

12
00:00:39,331 --> 00:00:42,316
to restore the water,
the roads, the electricity,

13
00:00:42,340 --> 00:00:45,456
which means then the construction people,
the insurance agents,

14
00:00:45,480 --> 00:00:48,094
all of them can get in
to rebuild the houses,

15
00:00:48,118 --> 00:00:51,106
which then means
you can restore the economy,

16
00:00:51,130 --> 00:00:56,003
and maybe even make it better
and more resilient to the next disaster.

17
00:00:57,899 --> 00:00:59,695
A major insurance company told me

18
00:00:59,719 --> 00:01:04,831
that if they can get a homeowner's claim
processed one day earlier,

19
00:01:04,855 --> 00:01:06,975
it'll make a difference of six months

20
00:01:06,999 --> 00:01:09,239
in that person getting
their home repaired.

21
00:01:10,118 --> 00:01:12,447
And that's why I do disaster robotics --

22
00:01:12,471 --> 00:01:16,970
because robots can
make a disaster go away faster.

23
00:01:18,305 --> 00:01:20,322
Now, you've already seen
a couple of these.

24
00:01:20,346 --> 00:01:22,066
These are the UAVs.

25
00:01:22,090 --> 00:01:23,951
These are two types of UAVs:

26
00:01:23,975 --> 00:01:25,840
a rotorcraft, or hummingbird;

27
00:01:25,864 --> 00:01:27,934
a fixed-wing, a hawk.

28
00:01:27,958 --> 00:01:31,348
And they're used extensively since 2005 --

29
00:01:31,372 --> 00:01:32,547
Hurricane Katrina.

30
00:01:32,571 --> 00:01:35,895
Let me show you how this hummingbird,
this rotorcraft, works.

31
00:01:35,919 --> 00:01:38,975
Fantastic for structural engineers.

32
00:01:38,999 --> 00:01:43,419
Being able to see damage from angles you
can't get from binoculars on the ground

33
00:01:43,443 --> 00:01:44,753
or from a satellite image,

34
00:01:44,777 --> 00:01:47,969
or anything flying at a higher angle.

35
00:01:48,530 --> 00:01:52,902
But it's not just structural engineers
and insurance people who need this.

36
00:01:52,926 --> 00:01:55,393
You've got things
like this fixed-wing, this hawk.

37
00:01:55,417 --> 00:01:58,688
Now, this hawk can be used
for geospatial surveys.

38
00:01:58,712 --> 00:02:01,561
That's where you're
pulling imagery together

39
00:02:01,585 --> 00:02:03,410
and getting 3D reconstruction.

40
00:02:03,434 --> 00:02:07,508
We used both of these at the Oso mudslides
up in Washington State,

41
00:02:07,532 --> 00:02:09,731
because the big problem

42
00:02:09,755 --> 00:02:12,913
was geospatial and hydrological
understanding of the disaster --

43
00:02:12,937 --> 00:02:14,183
not the search and rescue.

44
00:02:14,207 --> 00:02:16,526
The search and rescue teams
had it under control

45
00:02:16,550 --> 00:02:18,028
and knew what they were doing.

46
00:02:18,052 --> 00:02:21,922
The bigger problem was that river
and mudslide might wipe them out

47
00:02:21,946 --> 00:02:23,440
and flood the responders.

48
00:02:23,464 --> 00:02:27,206
And not only was it challenging
to the responders and property damage,

49
00:02:27,230 --> 00:02:30,509
it's also putting at risk
the future of salmon fishing

50
00:02:30,533 --> 00:02:32,249
along that part of Washington State.

51
00:02:32,273 --> 00:02:34,749
So they needed to understand
what was going on.

52
00:02:34,773 --> 00:02:37,382
In seven hours, going from Arlington,

53
00:02:37,406 --> 00:02:42,220
driving from the Incident Command Post
to the site, flying the UAVs,

54
00:02:42,244 --> 00:02:45,975
processing the data, driving back
to Arlington command post --

55
00:02:45,999 --> 00:02:47,405
seven hours.

56
00:02:47,429 --> 00:02:50,963
We gave them in seven hours
data that they could take

57
00:02:50,987 --> 00:02:54,737
only two to three days
to get any other way --

58
00:02:54,761 --> 00:02:56,598
and at higher resolution.

59
00:02:57,064 --> 00:02:58,437
It's a game changer.

60
00:02:59,773 --> 00:03:01,842
And don't just think about the UAVs.

61
00:03:01,866 --> 00:03:04,494
I mean, they are sexy -- but remember,

62
00:03:04,518 --> 00:03:07,854
80 percent of the world's
population lives by water,

63
00:03:07,878 --> 00:03:10,742
and that means our critical
infrastructure is underwater --

64
00:03:10,766 --> 00:03:14,052
the parts that we can't get to,
like the bridges and things like that.

65
00:03:14,076 --> 00:03:16,668
And that's why we have
unmanned marine vehicles,

66
00:03:16,692 --> 00:03:21,144
one type of which you've already met,
which is SARbot, a square dolphin.

67
00:03:21,168 --> 00:03:23,864
It goes underwater and uses sonar.

68
00:03:23,888 --> 00:03:26,221
Well, why are marine vehicles so important

69
00:03:26,245 --> 00:03:29,369
and why are they very, very important?

70
00:03:29,393 --> 00:03:30,909
They get overlooked.

71
00:03:30,933 --> 00:03:33,335
Think about the Japanese tsunami --

72
00:03:33,359 --> 00:03:37,325
400 miles of coastland totally devastated,

73
00:03:37,349 --> 00:03:42,045
twice the amount of coastland devastated
by Hurricane Katrina in the United States.

74
00:03:42,069 --> 00:03:45,925
You're talking about your bridges,
your pipelines, your ports -- wiped out.

75
00:03:45,949 --> 00:03:47,561
And if you don't have a port,

76
00:03:47,585 --> 00:03:50,831
you don't have a way
to get in enough relief supplies

77
00:03:50,855 --> 00:03:52,076
to support a population.

78
00:03:52,100 --> 00:03:54,672
That was a huge problem
at the Haiti earthquake.

79
00:03:55,565 --> 00:03:57,687
So we need marine vehicles.

80
00:03:57,711 --> 00:04:00,142
Now, let's look at a viewpoint
from the SARbot

81
00:04:00,166 --> 00:04:01,637
of what they were seeing.

82
00:04:01,661 --> 00:04:03,907
We were working on a fishing port.

83
00:04:03,931 --> 00:04:09,647
We were able to reopen that fishing port,
using her sonar, in four hours.

84
00:04:09,671 --> 00:04:12,324
That fishing port was told
it was going to be six months

85
00:04:12,348 --> 00:04:15,054
before they could get
a manual team of divers in,

86
00:04:15,078 --> 00:04:17,765
and it was going to take
the divers two weeks.

87
00:04:17,789 --> 00:04:20,083
They were going to miss
the fall fishing season,

88
00:04:20,107 --> 00:04:24,333
which was the major economy for that part,
which is kind of like their Cape Cod.

89
00:04:24,357 --> 00:04:26,625
UMVs, very important.

90
00:04:26,649 --> 00:04:29,897
But you know, all the robots
I've shown you have been small,

91
00:04:29,921 --> 00:04:33,698
and that's because robots
don't do things that people do.

92
00:04:33,722 --> 00:04:35,975
They go places people can't go.

93
00:04:35,999 --> 00:04:38,975
And a great example of that is Bujold.

94
00:04:38,999 --> 00:04:41,975
Unmanned ground vehicles
are particularly small,

95
00:04:41,999 --> 00:04:43,317
so Bujold --

96
00:04:43,341 --> 00:04:44,849
(Laughter)

97
00:04:44,873 --> 00:04:46,056
Say hello to Bujold.

98
00:04:46,080 --> 00:04:48,764
(Laughter)

99
00:04:49,606 --> 00:04:52,975
Bujold was used extensively
at the World Trade Center

100
00:04:52,999 --> 00:04:55,379
to go through Towers 1, 2 and 4.

101
00:04:55,403 --> 00:05:00,048
You're climbing into the rubble,
rappelling down, going deep in spaces.

102
00:05:00,072 --> 00:05:04,631
And just to see the World Trade Center
from Bujold's viewpoint, look at this.

103
00:05:04,655 --> 00:05:09,724
You're talking about a disaster
where you can't fit a person or a dog --

104
00:05:09,748 --> 00:05:11,890
and it's on fire.

105
00:05:11,914 --> 00:05:15,648
The only hope of getting
to a survivor way in the basement,

106
00:05:15,672 --> 00:05:17,867
you have to go through things
that are on fire.

107
00:05:17,891 --> 00:05:22,395
It was so hot, on one of the robots,
the tracks began to melt and come off.

108
00:05:23,161 --> 00:05:25,949
Robots don't replace people or dogs,

109
00:05:25,973 --> 00:05:28,494
or hummingbirds or hawks or dolphins.

110
00:05:28,518 --> 00:05:30,554
They do things new.

111
00:05:30,578 --> 00:05:36,267
They assist the responders,
the experts, in new and innovative ways.

112
00:05:36,291 --> 00:05:40,720
The biggest problem is not
making the robots smaller, though.

113
00:05:40,744 --> 00:05:42,975
It's not making them more heat-resistant.

114
00:05:42,999 --> 00:05:44,756
It's not making more sensors.

115
00:05:44,780 --> 00:05:48,390
The biggest problem is the data,
the informatics,

116
00:05:48,414 --> 00:05:52,450
because these people need to get
the right data at the right time.

117
00:05:52,474 --> 00:05:57,975
So wouldn't it be great if we could have
experts immediately access the robots

118
00:05:57,999 --> 00:06:00,975
without having to waste any time
of driving to the site,

119
00:06:00,999 --> 00:06:03,932
so whoever's there,
use their robots over the Internet.

120
00:06:03,956 --> 00:06:05,338
Well, let's think about that.

121
00:06:05,362 --> 00:06:08,734
Let's think about a chemical
train derailment in a rural county.

122
00:06:08,758 --> 00:06:12,600
What are the odds that the experts,
your chemical engineer,

123
00:06:12,624 --> 00:06:14,486
your railroad transportation engineers,

124
00:06:14,510 --> 00:06:19,371
have been trained on whatever UAV
that particular county happens to have?

125
00:06:19,395 --> 00:06:20,928
Probably, like, none.

126
00:06:20,952 --> 00:06:23,342
So we're using these kinds of interfaces

127
00:06:23,366 --> 00:06:27,706
to allow people to use the robots
without knowing what robot they're using,

128
00:06:27,730 --> 00:06:30,516
or even if they're using a robot or not.

129
00:06:32,252 --> 00:06:38,286
What the robots give you,
what they give the experts, is data.

130
00:06:38,310 --> 00:06:41,890
The problem becomes:
who gets what data when?

131
00:06:41,914 --> 00:06:45,807
One thing to do is to ship
all the information to everybody

132
00:06:45,831 --> 00:06:47,343
and let them sort it out.

133
00:06:47,367 --> 00:06:51,223
Well, the problem with that
is it overwhelms the networks,

134
00:06:51,247 --> 00:06:54,933
and worse yet, it overwhelms
the cognitive abilities

135
00:06:54,957 --> 00:06:59,466
of each of the people trying to get
that one nugget of information

136
00:06:59,490 --> 00:07:03,295
they need to make the decision
that's going to make the difference.

137
00:07:03,999 --> 00:07:07,093
So we need to think
about those kinds of challenges.

138
00:07:07,117 --> 00:07:08,382
So it's the data.

139
00:07:08,406 --> 00:07:10,642
Going back to the World Trade Center,

140
00:07:10,666 --> 00:07:14,975
we tried to solve that problem
by just recording the data from Bujold

141
00:07:14,999 --> 00:07:16,975
only when she was deep in the rubble,

142
00:07:16,999 --> 00:07:19,563
because that's what the USAR team
said they wanted.

143
00:07:20,523 --> 00:07:23,111
What we didn't know at the time

144
00:07:23,135 --> 00:07:25,733
was that the civil engineers
would have loved,

145
00:07:25,757 --> 00:07:29,975
needed the data as we recorded
the box beams, the serial numbers,

146
00:07:29,999 --> 00:07:32,618
the locations, as we went into the rubble.

147
00:07:33,343 --> 00:07:34,810
We lost valuable data.

148
00:07:34,834 --> 00:07:37,213
So the challenge is getting all the data

149
00:07:37,237 --> 00:07:39,417
and getting it to the right people.

150
00:07:39,441 --> 00:07:41,518
Now, here's another reason.

151
00:07:41,542 --> 00:07:43,853
We've learned that some buildings --

152
00:07:43,877 --> 00:07:47,149
things like schools,
hospitals, city halls --

153
00:07:47,173 --> 00:07:51,192
get inspected four times
by different agencies

154
00:07:51,216 --> 00:07:53,466
throughout the response phases.

155
00:07:54,156 --> 00:07:57,425
Now, we're looking, if we can get
the data from the robots to share,

156
00:07:57,449 --> 00:08:02,299
not only can we do things like
compress that sequence of phases

157
00:08:02,323 --> 00:08:03,976
to shorten the response time,

158
00:08:04,000 --> 00:08:08,358
but now we can begin
to do the response in parallel.

159
00:08:08,382 --> 00:08:09,761
Everybody can see the data.

160
00:08:09,785 --> 00:08:11,247
We can shorten it that way.

161
00:08:11,999 --> 00:08:15,405
So really, "disaster robotics"
is a misnomer.

162
00:08:16,227 --> 00:08:18,122
It's not about the robots.

163
00:08:18,146 --> 00:08:20,028
It's about the data.

164
00:08:20,433 --> 00:08:23,763
(Applause)

165
00:08:23,787 --> 00:08:25,744
So my challenge to you:

166
00:08:25,768 --> 00:08:28,068
the next time you hear about a disaster,

167
00:08:28,092 --> 00:08:29,417
look for the robots.

168
00:08:29,441 --> 00:08:32,566
They may be underground,
they may be underwater,

169
00:08:32,590 --> 00:08:34,284
they may be in the sky,

170
00:08:34,308 --> 00:08:36,273
but they should be there.

171
00:08:36,297 --> 00:08:37,464
Look for the robots,

172
00:08:37,488 --> 00:08:40,235
because robots are coming to the rescue.

173
00:08:40,259 --> 00:08:45,920
(Applause)