WEBVTT

00:00:00.000 --> 00:00:04.037
Who cares? >> That's true. The human's
not in the thing makes it unsafe.

00:00:04.120 --> 00:00:07.877
>> So, a lot of pilots, they're not trying
to do the maximally safe thing. They I'm

00:00:07.960 --> 00:00:11.677
not saying pilots are unsafe.
Remember that pilots are working within

00:00:11.760 --> 00:00:13.477
industry constraints. Like, a guy who

00:00:13.560 --> 00:00:16.077
flies a bizjet or a charter helicopter

00:00:16.160 --> 00:00:20.997
operation, he is flying within the same
constraints as a chauffeur, right? He

00:00:21.080 --> 00:00:22.637
You got to approach everything smooth.

00:00:22.720 --> 00:00:26.357
You If something's going wrong,
you don't want it to show until it has to,

00:00:26.440 --> 00:00:30.157
right? You You are not going to want
to take aggressive really aggressive

00:00:30.240 --> 00:00:33.037
evasive maneuvers at the first sign
of trouble. If I'm a robot, that's what

00:00:33.120 --> 00:00:36.517
it's going to do. It's going to see
the first sign of possible trouble and do

00:00:36.600 --> 00:00:40.677
the most evasive action. Your charter
helicopter pilot is not going to do that

00:00:40.760 --> 00:00:42.597
because you're going to get really bad
reviews and you're going to lose your

00:00:42.680 --> 00:00:45.597
job really fast. I think just the way
that you look at these things are very

00:00:45.680 --> 00:00:48.997
different between human pilots and
unmanned pilots. My experience has been

00:00:49.080 --> 00:00:53.237
that a lot of like let's take like
fighter pilots, for example. Yes, they

00:00:53.320 --> 00:00:55.757
love to fly, but more they want to win.

00:00:55.840 --> 00:00:57.877
>> Well, I think I think with our
autonomous systems, it really was like

00:00:57.960 --> 00:01:01.117
seeing is believing, right? Like,
we demonstrated the capability and now

00:01:01.200 --> 00:01:02.637
everyone's on board.
The collision avoidance.

00:01:02.720 --> 00:01:05.637
>> Well, you run the missions and you
run the mission simulations. And also with

00:01:05.720 --> 00:01:08.837
human pilots, you run them in simulators.
You're like, "Okay, let's

00:01:08.920 --> 00:01:12.797
see you not hit the terrain in this
situation, right? Your GPS is out. Look

00:01:12.880 --> 00:01:16.917
at this recent GPS jamming scenario where
where it took out the medevac pilots."

00:01:17.000 --> 00:01:20.797
Like, so put a lot of pilots into
that same situation and they're going to

00:01:20.880 --> 00:01:23.997
they're going to have a they're going
to controlled flight into terrain

00:01:24.080 --> 00:01:26.837
for sure. And maybe a few of them will
avoid it. And then you integrate that

00:01:26.920 --> 00:01:30.117
with some of these safety systems and all
of a sudden every person in the sim

00:01:30.200 --> 00:01:33.037
lives. And I think that does convince
people. We need to show people more of

00:01:33.120 --> 00:01:35.317
that stuff. >> So, I got
a follow-on now. I've seen

00:01:35.400 --> 00:01:38.317
things that work great in the sim, they
work great in a laboratory environment.

00:01:38.400 --> 00:01:41.637
And this is not just autonomy, this can
be sensors. I've done a lot of helmet

00:01:41.720 --> 00:01:45.157
queuing type of things. Work great
in the sim, you get in the real flight

00:01:45.240 --> 00:01:47.877
environment, you get to some of those
edge cases, it's a lot more challenging.

00:01:47.960 --> 00:01:49.357
>> You need to have
exercises that are that

00:01:49.440 --> 00:01:53.677
are realistic and representative of your
actual workflow and workload and then

00:01:53.760 --> 00:01:57.517
ideally overloaded. One of the big flaws
of a lot of exercises, and including

00:01:57.600 --> 00:02:01.357
ones that we're part of is that they
have this nice scripted thing and here's

00:02:01.440 --> 00:02:04.437
how it's all going to go and it's all
very easy. Like you have a bunch of

00:02:04.520 --> 00:02:06.797
beyond visual line of sight engagements.

00:02:06.880 --> 00:02:09.557
Oh, we've picked it up.
We passed the queue and my

00:02:09.640 --> 00:02:10.917
I've got I've got to fire my weapon.

00:02:11.000 --> 00:02:12.637
It's going to pick up the track.

00:02:12.720 --> 00:02:16.637
Great, it all works. Well, what happens
when actually there was a bunch of stuff

00:02:16.720 --> 00:02:20.197
that you didn't expect and now it
is shooting at you. Also, a bunch of your

00:02:20.280 --> 00:02:23.357
support systems are offline.
Maybe you've actually lost some of your

00:02:23.440 --> 00:02:27.357
stability augmentation system in your
rotorcraft. Maybe because it was shot or

00:02:27.440 --> 00:02:30.437
maybe just because the combat cadence
maybe we're in the middle of the third

00:02:30.520 --> 00:02:33.637
week of the war and you haven't been
back to actually get properly maintained

00:02:33.720 --> 00:02:37.037
in 3 weeks and so it's just total
and not working anymore. Like now what

00:02:37.120 --> 00:02:40.757
is it like to do it? Or one of my favorite
ones like I most my daily flyer

00:02:40.840 --> 00:02:43.757
is a Eurocopter EC120. I love I love it.

00:02:43.840 --> 00:02:46.517
I know it's a little underpowered
and I'd rather be flying an MD500 but

00:02:46.600 --> 00:02:47.597
there's things that I like about it.

00:02:47.680 --> 00:02:50.357
>> [laughter] >> Like I
can fly a hydraulics out.

00:02:50.440 --> 00:02:54.677
>> Mhm. >> I really would not want to be
flying a hydraulics out in a high workload

00:02:54.760 --> 00:02:57.037
situation with a lot of other stuff.

00:02:57.120 --> 00:02:58.317
>> in the weather flying at night.

00:02:58.400 --> 00:03:02.357
>> night in the weather or and it worse
maybe at night in the weather and I

00:03:02.440 --> 00:03:06.037
didn't even expect to be at night in
the weather. Right? So or worse, I end up

00:03:06.120 --> 00:03:08.717
overseas and I'm not wearing
my emergency suit and I'm thinking, "Okay,

00:03:08.800 --> 00:03:11.957
what risk do I want to take? How far
like I want to be able to auto rotate to

00:03:12.040 --> 00:03:16.157
the coast if I lose my main power. Maybe
my chip light is on. I know that I've

00:03:16.240 --> 00:03:19.877
only got so much time left on my engine."
Unexpected scenarios should be

00:03:19.960 --> 00:03:23.757
part of these exercises. We should be We
should be putting people into the worst

00:03:23.840 --> 00:03:27.397
case scenarios and then seeing how this
stuff works out. And I'll be honest, a

00:03:27.480 --> 00:03:30.277
lot of our stuff totally fails to work
when you get in those scenarios because

00:03:30.360 --> 00:03:32.997
my guys are computer kids, right?
And like we work we have a lot of veterans

00:03:33.080 --> 00:03:35.597
here and we try to work custom with the
customers to understand these problems

00:03:35.680 --> 00:03:38.957
but it's very easy to develop a system
that works great in the science fair

00:03:39.040 --> 00:03:42.077
project scenario and then you put it
into something real and it all falls

00:03:42.160 --> 00:03:45.197
apart. And so I >> This is why
I'm all about early test. I

00:03:45.280 --> 00:03:48.077
early test as soon as we can when we're
in the design phase so that we can find

00:03:48.160 --> 00:03:49.237
the problems and fix the problems.

00:03:49.320 --> 00:03:52.757
>> That's right. If you wait too long,
sometimes the fixes are just totally

00:03:52.840 --> 00:03:56.237
non-viable. Like to use an extreme
example, what if you were considering

00:03:56.320 --> 00:03:59.877
voice controls and it reads out over
the radio and everyone gets to hear it

00:03:59.960 --> 00:04:03.357
versus a dedicated tablet for controlling
something. What if you find

00:04:03.440 --> 00:04:08.157
out that the tablet's just completely
unworkable in high workload environments?

00:04:08.240 --> 00:04:12.117
The only fix at that point is try
to make the UI on the tablet better. It's

00:04:12.200 --> 00:04:15.477
very hard to say, "Well, that's was 5
years of development down the drain.

00:04:15.560 --> 00:04:17.837
We're just going to rip this out and start
from >> Well, touchscreen is one of those

00:04:17.920 --> 00:04:19.917
examples touchscreen
can be challenging and

00:04:20.000 --> 00:04:21.357
>> I hate touchscreens. They're so bad.

00:04:21.440 --> 00:04:25.277
They're so bad. Don't even get me started.
Steve Jobs, I I like the guy,

00:04:25.360 --> 00:04:27.557
but man, he ruined technology.

00:04:27.640 --> 00:04:30.197
>> I miss my Blackberry. >> Absolutely.

00:04:30.280 --> 00:04:33.277
>> I love tactile feedback. I really do.
I have a whole >> That's a huge thing.

00:04:33.360 --> 00:04:35.917
>> It's so critical. >> It's
the tactile feedback. That's like a

00:04:36.000 --> 00:04:38.797
very specific example of what I'm
talking about. I've got this whole

00:04:38.880 --> 00:04:43.960
anti-Steve Jobs and I remember
I respect Steve Jobs to death.