0:00:00.999,0:00:05.975
Over a million people are killed[br]each year in disasters.

0:00:05.999,0:00:11.233
Two and a half million people[br]will be permanently disabled or displaced,

0:00:11.257,0:00:15.181
and the communities will take[br]20 to 30 years to recover

0:00:15.205,0:00:17.606
and billions of economic losses.

0:00:19.054,0:00:23.425
If you can reduce[br]the initial response by one day,

0:00:23.449,0:00:26.976
you can reduce the overall recovery

0:00:27.000,0:00:29.808
by a thousand days, or three years.

0:00:29.832,0:00:31.585
See how that works?

0:00:31.609,0:00:34.459
If the initial responders[br]can get in, save lives,

0:00:34.483,0:00:36.999
mitigate whatever flooding[br]danger there is,

0:00:37.023,0:00:39.307
that means the other groups can get in

0:00:39.331,0:00:42.316
to restore the water,[br]the roads, the electricity,

0:00:42.340,0:00:45.456
which means then the construction people,[br]the insurance agents,

0:00:45.480,0:00:48.094
all of them can get in[br]to rebuild the houses,

0:00:48.118,0:00:51.106
which then means[br]you can restore the economy,

0:00:51.130,0:00:56.003
and maybe even make it better[br]and more resilient to the next disaster.

0:00:57.899,0:00:59.695
A major insurance company told me

0:00:59.719,0:01:04.831
that if they can get a homeowner's claim[br]processed one day earlier,

0:01:04.855,0:01:06.975
it'll make a difference of six months

0:01:06.999,0:01:09.239
in that person getting[br]their home repaired.

0:01:10.118,0:01:12.447
And that's why I do disaster robotics --

0:01:12.471,0:01:16.970
because robots can[br]make a disaster go away faster.

0:01:18.305,0:01:20.322
Now, you've already seen[br]a couple of these.

0:01:20.346,0:01:22.066
These are the UAVs.

0:01:22.090,0:01:23.951
These are two types of UAVs:

0:01:23.975,0:01:25.840
a rotorcraft, or hummingbird;

0:01:25.864,0:01:27.934
a fixed-wing, a hawk.

0:01:27.958,0:01:31.348
And they're used extensively since 2005 --

0:01:31.372,0:01:32.547
Hurricane Katrina.

0:01:32.571,0:01:35.895
Let me show you how this hummingbird,[br]this rotorcraft, works.

0:01:35.919,0:01:38.975
Fantastic for structural engineers.

0:01:38.999,0:01:43.419
Being able to see damage from angles you[br]can't get from binoculars on the ground

0:01:43.443,0:01:44.753
or from a satellite image,

0:01:44.777,0:01:47.969
or anything flying at a higher angle.

0:01:48.530,0:01:52.902
But it's not just structural engineers[br]and insurance people who need this.

0:01:52.926,0:01:55.393
You've got things[br]like this fixed-wing, this hawk.

0:01:55.417,0:01:58.688
Now, this hawk can be used[br]for geospatial surveys.

0:01:58.712,0:02:01.561
That's where you're[br]pulling imagery together

0:02:01.585,0:02:03.410
and getting 3D reconstruction.

0:02:03.434,0:02:07.508
We used both of these at the Oso mudslides[br]up in Washington State,

0:02:07.532,0:02:09.731
because the big problem

0:02:09.755,0:02:12.913
was geospatial and hydrological[br]understanding of the disaster --

0:02:12.937,0:02:14.183
not the search and rescue.

0:02:14.207,0:02:16.526
The search and rescue teams[br]had it under control

0:02:16.550,0:02:18.028
and knew what they were doing.

0:02:18.052,0:02:21.922
The bigger problem was that river[br]and mudslide might wipe them out

0:02:21.946,0:02:23.440
and flood the responders.

0:02:23.464,0:02:27.206
And not only was it challenging[br]to the responders and property damage,

0:02:27.230,0:02:30.509
it's also putting at risk[br]the future of salmon fishing

0:02:30.533,0:02:32.249
along that part of Washington State.

0:02:32.273,0:02:34.749
So they needed to understand[br]what was going on.

0:02:34.773,0:02:37.382
In seven hours, going from Arlington,

0:02:37.406,0:02:42.220
driving from the Incident Command Post[br]to the site, flying the UAVs,

0:02:42.244,0:02:45.975
processing the data, driving back[br]to Arlington command post --

0:02:45.999,0:02:47.405
seven hours.

0:02:47.429,0:02:50.963
We gave them in seven hours[br]data that they could take

0:02:50.987,0:02:54.737
only two to three days[br]to get any other way --

0:02:54.761,0:02:56.598
and at higher resolution.

0:02:57.064,0:02:58.437
It's a game changer.

0:02:59.773,0:03:01.842
And don't just think about the UAVs.

0:03:01.866,0:03:04.494
I mean, they are sexy -- but remember,

0:03:04.518,0:03:07.854
80 percent of the world's[br]population lives by water,

0:03:07.878,0:03:10.742
and that means our critical[br]infrastructure is underwater --

0:03:10.766,0:03:14.052
the parts that we can't get to,[br]like the bridges and things like that.

0:03:14.076,0:03:16.668
And that's why we have[br]unmanned marine vehicles,

0:03:16.692,0:03:21.144
one type of which you've already met,[br]which is SARbot, a square dolphin.

0:03:21.168,0:03:23.864
It goes underwater and uses sonar.

0:03:23.888,0:03:26.221
Well, why are marine vehicles so important

0:03:26.245,0:03:29.369
and why are they very, very important?

0:03:29.393,0:03:30.909
They get overlooked.

0:03:30.933,0:03:33.335
Think about the Japanese tsunami --

0:03:33.359,0:03:37.325
400 miles of coastland totally devastated,

0:03:37.349,0:03:42.045
twice the amount of coastland devastated[br]by Hurricane Katrina in the United States.

0:03:42.069,0:03:45.925
You're talking about your bridges,[br]your pipelines, your ports -- wiped out.

0:03:45.949,0:03:47.561
And if you don't have a port,

0:03:47.585,0:03:50.831
you don't have a way[br]to get in enough relief supplies

0:03:50.855,0:03:52.076
to support a population.

0:03:52.100,0:03:54.672
That was a huge problem[br]at the Haiti earthquake.

0:03:55.565,0:03:57.687
So we need marine vehicles.

0:03:57.711,0:04:00.142
Now, let's look at a viewpoint[br]from the SARbot

0:04:00.166,0:04:01.637
of what they were seeing.

0:04:01.661,0:04:03.907
We were working on a fishing port.

0:04:03.931,0:04:09.647
We were able to reopen that fishing port,[br]using her sonar, in four hours.

0:04:09.671,0:04:12.324
That fishing port was told[br]it was going to be six months

0:04:12.348,0:04:15.054
before they could get[br]a manual team of divers in,

0:04:15.078,0:04:17.765
and it was going to take[br]the divers two weeks.

0:04:17.789,0:04:20.083
They were going to miss[br]the fall fishing season,

0:04:20.107,0:04:24.333
which was the major economy for that part,[br]which is kind of like their Cape Cod.

0:04:24.357,0:04:26.625
UMVs, very important.

0:04:26.649,0:04:29.897
But you know, all the robots[br]I've shown you have been small,

0:04:29.921,0:04:33.698
and that's because robots[br]don't do things that people do.

0:04:33.722,0:04:35.975
They go places people can't go.

0:04:35.999,0:04:38.975
And a great example of that is Bujold.

0:04:38.999,0:04:41.975
Unmanned ground vehicles[br]are particularly small,

0:04:41.999,0:04:43.317
so Bujold --

0:04:43.341,0:04:44.849
(Laughter)

0:04:44.873,0:04:46.056
Say hello to Bujold.

0:04:46.080,0:04:48.764
(Laughter)

0:04:49.606,0:04:52.975
Bujold was used extensively[br]at the World Trade Center

0:04:52.999,0:04:55.379
to go through Towers 1, 2 and 4.

0:04:55.403,0:05:00.048
You're climbing into the rubble,[br]rappelling down, going deep in spaces.

0:05:00.072,0:05:04.631
And just to see the World Trade Center[br]from Bujold's viewpoint, look at this.

0:05:04.655,0:05:09.724
You're talking about a disaster[br]where you can't fit a person or a dog --

0:05:09.748,0:05:11.890
and it's on fire.

0:05:11.914,0:05:15.648
The only hope of getting[br]to a survivor way in the basement,

0:05:15.672,0:05:17.867
you have to go through things[br]that are on fire.

0:05:17.891,0:05:22.395
It was so hot, on one of the robots,[br]the tracks began to melt and come off.

0:05:23.161,0:05:25.949
Robots don't replace people or dogs,

0:05:25.973,0:05:28.494
or hummingbirds or hawks or dolphins.

0:05:28.518,0:05:30.554
They do things new.

0:05:30.578,0:05:36.267
They assist the responders,[br]the experts, in new and innovative ways.

0:05:36.291,0:05:40.720
The biggest problem is not[br]making the robots smaller, though.

0:05:40.744,0:05:42.975
It's not making them more heat-resistant.

0:05:42.999,0:05:44.756
It's not making more sensors.

0:05:44.780,0:05:48.390
The biggest problem is the data,[br]the informatics,

0:05:48.414,0:05:52.450
because these people need to get[br]the right data at the right time.

0:05:52.474,0:05:57.975
So wouldn't it be great if we could have[br]experts immediately access the robots

0:05:57.999,0:06:00.975
without having to waste any time[br]of driving to the site,

0:06:00.999,0:06:03.932
so whoever's there,[br]use their robots over the Internet.

0:06:03.956,0:06:05.338
Well, let's think about that.

0:06:05.362,0:06:08.734
Let's think about a chemical[br]train derailment in a rural county.

0:06:08.758,0:06:12.600
What are the odds that the experts,[br]your chemical engineer,

0:06:12.624,0:06:14.486
your railroad transportation engineers,

0:06:14.510,0:06:19.371
have been trained on whatever UAV[br]that particular county happens to have?

0:06:19.395,0:06:20.928
Probably, like, none.

0:06:20.952,0:06:23.342
So we're using these kinds of interfaces

0:06:23.366,0:06:27.706
to allow people to use the robots[br]without knowing what robot they're using,

0:06:27.730,0:06:30.516
or even if they're using a robot or not.

0:06:32.252,0:06:38.286
What the robots give you,[br]what they give the experts, is data.

0:06:38.310,0:06:41.890
The problem becomes:[br]who gets what data when?

0:06:41.914,0:06:45.807
One thing to do is to ship[br]all the information to everybody

0:06:45.831,0:06:47.343
and let them sort it out.

0:06:47.367,0:06:51.223
Well, the problem with that[br]is it overwhelms the networks,

0:06:51.247,0:06:54.933
and worse yet, it overwhelms[br]the cognitive abilities

0:06:54.957,0:06:59.466
of each of the people trying to get[br]that one nugget of information

0:06:59.490,0:07:03.295
they need to make the decision[br]that's going to make the difference.

0:07:03.999,0:07:07.093
So we need to think[br]about those kinds of challenges.

0:07:07.117,0:07:08.382
So it's the data.

0:07:08.406,0:07:10.642
Going back to the World Trade Center,

0:07:10.666,0:07:14.975
we tried to solve that problem[br]by just recording the data from Bujold

0:07:14.999,0:07:16.975
only when she was deep in the rubble,

0:07:16.999,0:07:19.563
because that's what the USAR team[br]said they wanted.

0:07:20.523,0:07:23.111
What we didn't know at the time

0:07:23.135,0:07:25.733
was that the civil engineers[br]would have loved,

0:07:25.757,0:07:29.975
needed the data as we recorded[br]the box beams, the serial numbers,

0:07:29.999,0:07:32.618
the locations, as we went into the rubble.

0:07:33.343,0:07:34.810
We lost valuable data.

0:07:34.834,0:07:37.213
So the challenge is getting all the data

0:07:37.237,0:07:39.417
and getting it to the right people.

0:07:39.441,0:07:41.518
Now, here's another reason.

0:07:41.542,0:07:43.853
We've learned that some buildings --

0:07:43.877,0:07:47.149
things like schools,[br]hospitals, city halls --

0:07:47.173,0:07:51.192
get inspected four times[br]by different agencies

0:07:51.216,0:07:53.466
throughout the response phases.

0:07:54.156,0:07:57.425
Now, we're looking, if we can get[br]the data from the robots to share,

0:07:57.449,0:08:02.299
not only can we do things like[br]compress that sequence of phases

0:08:02.323,0:08:03.976
to shorten the response time,

0:08:04.000,0:08:08.358
but now we can begin[br]to do the response in parallel.

0:08:08.382,0:08:09.761
Everybody can see the data.

0:08:09.785,0:08:11.247
We can shorten it that way.

0:08:11.999,0:08:15.405
So really, "disaster robotics"[br]is a misnomer.

0:08:16.227,0:08:18.122
It's not about the robots.

0:08:18.146,0:08:20.028
It's about the data.

0:08:20.433,0:08:23.763
(Applause)

0:08:23.787,0:08:25.744
So my challenge to you:

0:08:25.768,0:08:28.068
the next time you hear about a disaster,

0:08:28.092,0:08:29.417
look for the robots.

0:08:29.441,0:08:32.566
They may be underground,[br]they may be underwater,

0:08:32.590,0:08:34.284
they may be in the sky,

0:08:34.308,0:08:36.273
but they should be there.

0:08:36.297,0:08:37.464
Look for the robots,

0:08:37.488,0:08:40.235
because robots are coming to the rescue.

0:08:40.259,0:08:45.920
(Applause)