you and I rely on the ability to
communicate every day right we do it to
understand the world around us and to
understand how we operate today I want
you to start by thinking about those
people around you with you communicate
with right
your parents your spouse your friends
right maybe the police or a physician
and I want you to think about the type
of information you communicate with them
today I like to present someone to you
with whom I communicate who explores
around around himself and lets me know
about what he’s experiencing but unlike
what you might be imagining right this
special someone is Goodwin
now Goodwin is not hiding behind the
bushes Goodwin is the building a
five-story 160,000 square foot building
right but Goodwin is a little bit
special Goodwin has the ability to sense
in the and he’s starting to communicate
with us to tell us things about himself
as a matter of fact he senses things
every day he think he senses crowds of
people coming into work he sends his
crowds of people going out to lunch he
senses that some of them come back from
lunch right he feels the weather he
feels the rain he fills the wind right
so this is a little bit different from
what you might expect from a building so
to get us started I like to show you an
example on September 7 2016 just a few
weeks ago Pawnee Oklahoma experienced a
5.8 earthquake the earthquake right and
it’s ripples spread across the country
like ripples in a pond but when it came
to Tennessee in Virginia we didn’t feel
it anymore
but unlike you and I Goodwin actually
felt it these are the sensors in Goodwin
showing you that about 8:10 a.m. on that
Saturday morning right the building
started to move so we could say that 909
miles away a hundred and eighty seven
seconds later Pawnee met Goodwin and
Goodwin was all shook up right so what
you see here right
is an animation of what the censors were
feeling through that moment through that
earthquake and although it wasn’t that
bad as the picture shows right it’s been
exaggerated for visual purposes so we
can understand the motion in the
building it lets you know that Goodwin
could actually start to communicate with
us of what he’s feeling in what he’s
going through right so with that in mind
I like to ask you a question right what
has your building done for you lately
right I know that’s kind of weird right
but what has your building done for you
lately has it been able to adapt right
in order to save you energy has it
tracked your health has it kept you safe
as it helped your grandmother and your
grandfather in that fall you might these
things you might be thinking these are
these are not normal questions but
that’s why we’re here all right so with
that I like to take you on my journey
what started out as a spark a small idea
and has led us today to one of the most
instrumented buildings in the world for
vibration this is this is changing for
us how we think of a lot of building how
we interact with the building right and
what we can start to do in the near
future hopefully today I can challenge
your preconceived notion of a building
as this inanimate object that doesn’t do
anything for you right and together I
can show you how we which is the
important word in the sentence are
changing how this is done and how it
takes a multidisciplinary group to take
this and do it so let’s back up a little
bit and start from the beginning so how
did we get started on the fall of 2011 I
was hired at Virginia Tech in the
mechanical engineering department and I
had just come from England from putting
hundreds of accelerometers on
helicopters and I said across my office
and this was the view from my office
it was Goodwin Hall being built back
then called the signature engineering
building and I decided that it would be
a good idea to place these sensors
throughout the building at a very large
scale what’s funny about ideas is
actually how they flourish and that they
need to be nourished
so the oddest thing was that I went to a
soccer game I ran into one of the
associate Dean’s of the College of
Engineering and we found out that
together we had this idea that we wanted
to make Goodwin a very special building
so then the next thing I did was what
any professor does grabs his best assets
right his students
and they’re always willing to do all
kinds of crazy things with us and we
went out and started to collect pilot
data putting sensors measuring scene if
we can get anything out of the building
that would make any sense we started to
share that idea with other faculty
members who very quickly bought into the
idea and had the ideas of their own of
what we could do with this data right
three long and hardworking years later
we had we had instrumented Goodwin home
to be the most the most instrument of
building the world for vibrations with
213 sensors throughout the entire
structure sixty-five thousand feet of
cable connecting every sensor to every
computer that collects data and it just
so happens when you buy sixty five
thousand feet of cable you can get it at
any color you want and hookie orange is
one of them so that was great so the
next thing we did was we pulled the
switch and the data started to come in
and the question was then what do we do
with this data so much like
Frankenstein’s monster right by putting
all these sensors throughout the
building
we had just given the building a
heartbeat and now we were monitoring
this heartbeat but 213 of these examples
today we’ve learned that what you’re
seeing right there are serious F
multiple events a door is opening and
closing a person is walking across the
hallway where there’s a sensor and a
piece of equipment machinery is turning
on right so then we set out to say okay
what can we find out and what can we do
and how can we use this to better our
society so you clearly saw the first
example which was the earthquake which
actually shook the building but
buildings move in very peculiar motions
structures in themselves with this we
were learning that the models we make
can be better can be better tailored we
can make more efficient buildings we can
make safer buildings and this level of
understanding is changing the way we can
understand buildings but like any good
doctor we wanted to go deeper inside
what you see here is the motion of the
individual sensors letting us know
what’s going on right we’re yellow is is
a sensor that is not seeing a lot of
activity and red is a sensor that’s seen
a lot of activity you can see that as
the day goes by more
more the sensors turn red which
correlate really well with actually
class exchanges lunch times and so on
and this and now it starts to be an idea
of what you can do with this information
right we can understand how to control
HVAC better we can control lighting
better we can save a lot of energy when
it comes to a building but we could also
help people during an emergency
evacuation so when we saw this
information we wanted to go even in
deeper so what is costing the sensors to
actually respond this way well it turns
out that very small vibrations in the
building which could be caused by the
heel strike of your walking right span
through the floor like ripples like the
earthquake and our sensors start to pick
them up and what you see there is a
simulation of these vibrations
throughout the building right and much
like you’re finding me on stage you’re
using both of your ears to triangulate
where I am we can use that information
the same way to find and locate people
in the building we can do it today under
very controlled experiments and we can
find a person in a hundred sixty
thousand square foot building within a
one foot radius right so okay so we said
okay we’re starting to understand how
the building reacts but can we start to
understand what is causing the incidents
to occur and the next thing we did is we
wanted to study people what you see here
is in red a woman walking across the
hallway with high heels and in blue a
male walking across the hallway with
soft shoes and you and I can clearly
tell the difference between those two
but what we’re trying to do today is get
computers to do this for us to develop
mathematical models that are capable of
distinguishing this but we didn’t want
to do it this easy so what we obviously
did to the computers is we gave them a
little bit more of a challenge we mix
the shoes up we had everybody wear
Horsell shoes and soft sole shoes and as
of today we can distinguish gender with
a 97% accuracy right you might be
starting to freak out and okay that’s
fine
we don’t like to do things safe all the
time but you can imagine that if we’re
searching for someone and we can divide
the classification by 50% our ability to
find someone reduces right and we can
save people much quicker if we’re
looking for some right some other people
are interested in things like customized
marketing in the mall so that as you
walk we can feed you exactly what you
want we want you to buy right that’s up
to them anyway so of course this leads
to a lot of questions of privacy and
just like today you left your
fingerprints here in this wonderful
auditorium you’re not worried because I
don’t have a database to pull you out
and compare that fingerprint to that
base so you can see how this technology
can actually be scaled for privacy it
won’t necessarily be an issue but with
that in mind you know weird experiments
have not always been received well and
this is not the first time actually a
very famous dr. Frankenstein in the film
of 1931 his advice was said to him right
you have created a monster and he will
destroy you and I would hope that we
would react a little bit better and move
towards Frankenstein’s answer and he
says patience patience I believe in this
monster as you call it right and if you
don’t well leave all right so far he’s
been kept in complete darkness wait till
I bring him into the light so hopefully
today right now with the next few
examples I’ll show you some light let’s
start with some interesting statistics
56 percent of older adults who fall die
within six months results in 2.3 million
visits to the hospital 20,000 deaths a
year right and 30 billion dollars in
cost what if the building could actually
let us know that it’s happened
automatically and deploy emergency to
that situation this research in video is
quite hard to do because for some reason
it’s not okay to invite people to the
building and triple so simulating this
day that requires a little bit of
finesse which we haven’t gotten there
yet but my students are developing some
fun ways of trying to figure this out
sometimes they need a little push
um anyway but what I want you to get out
of this is that what the building sees
when an event like this happens in the
previous ones we’ve talked about a few
things are very different so the top one
is a simulation of a dummy falling right
the bottom one is the door closing so if
we can develop mathematical models that
are able to pull this apart at all time
run 24/7 right there’s a lot of
information to process and you can see
how this is not just the work of a
mechanical engineer right computer
science and mathematics and so on so
it’s a very very tough problem so with
that I like you to imagine the future
with me of say we couldn’t lead do this
all at the same time this is a floor
plan of the fourth floor with Goodwin
Hall so for example let’s imagine that
the building detects the situation right
at that moment it moves forward it says
I’ve located this event
I’m classifying it I’m trying to
figuring out what it is so I’m deploying
countermeasures right at that point it
might tell you on your phone or in your
watch what’s going on right it might
then decide to call the police and say
hey we have an intruder in Goodwin Hall
right he’s in the third floor he’s
moving he’s male he weighs 190 pounds
right he’s discharging 9-millimeter
weapon and I’m currently tracking him
right the police could then respond and
say we’re on our way
please help these people out you say
okay I’ve locked the third floor I’m
evacuating one two and four right the
building can then respond to the police
the police arrives the building deploys
that information to their portable
devices it tells them what is the
fastest route to enter the building
right it tracks them coming into the
building and then it gives them
second-by-second information about their
approach right the building could be the
first responder and in a lot of
situations that’s all the time we have
right so today hopefully you’ve been a
little bit challenged about your
preconceived notion of what a building
might do in the future
right how a small idea when its
nourished right can flourish into
something quite unique how it takes a
multidisciplinary team to get these
ideas carried out today so for me that’s
why I think it takes a smart village
thank you you