Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Some task I think we should dothrough robotic work for something which is not
safe, which is really heavy.That's how gray body unhealthy. We will
not replace a human, but wewill enable and empower a human worker.
Today I speak with Maricha Krizla,who heads a Siemens research group in Berkeley,
(00:21):
California. Her smart Machine vision andvisualization team focuses on augmented reality and
how it can be used to helphumans and technology work better together, often
by giving people what she likes tocall digital superpowers. Today we talk about
how her background and geoinformatics led toher fascinating research involving mice, r f
I D chips, and Alzheimer's.We'll learn how her current work is transforming
(00:44):
augmented reality from an entertaining gimmick intoa powerful tool to make an impact on
society by improving workplace safety and evenhelping scientists protect our coral reefs. Here's
my conversation with Maricha Krizla. Welive in a time where design and technology
(01:06):
touch every aspect of our lives.But where did it all come from?
Who designed it, how is itbuilt and brought to market? What will
it look like in a year?Two years, a hundred years from the
phones and smart watches that help usin our day to day to the cutting
edge spaceships and three D printers thatare leading us into the future. Modern
design is constantly shaping the way wework, communicate, problem solve, and
play, and every new design,bigger, small, starts with an idea
(01:29):
and a bill of materials. I'magenta Strongheart, and this is the Bomb
where we talk to leading innovators inthe tech world and celebrate the transformational power
of design. Welcome to the Bomb. I'm so happy to have you here
today, and it is great tofirst get introduced to your lab a couple
(01:51):
of days ago and visit your spacein Berkeley. I think it'd be helpful
probably for our listeners to just kindof start with your current role and what
you're doing your lab. Thank youfor having me. I'm a research group
hat for a Semis technology. Weare located in Berkeley. My team is
split between the coasts and we doresearch for smart machine vision and visitalization that
(02:13):
I've imagine covers a lot of ground, especially these days, there's so much
going on, so much technology evolvingvery quickly in those spaces, so I'm
really excited to dig into that furtherin this conversation. I got to see
a few projects when I was there, but I think that it might be
helpful to get a little bit ofa grounding of how long you've been in
(02:34):
this space, how long you've beenwith Semens Technology before we kind of rewind
and learn more about how you gotthere. So you've been there over ten
years, I believe, right,and not quite yet, but I'm going
close. So I've been with SemensTechnology always Semi Technology in Berkeley for around
nine years. I think my ninthunibushry is May fifth. Okay, awesome.
And we have this office in Berkeleywell, which we share with one
(02:58):
other research group. And yeah,we do research for all of our different
Semens businesses. And so could youwalk us through the the core kind of
Semens businesses that you guys serve.So this will be all of the Semens
businesses. We have Semens doing mobility, Semens Mobility. We have the health
and news, which are now alittle bit separate. We have Semens Digital
(03:22):
Industries, Semens Digital factories, soobviously from factory automation, healthcare, mobility,
and Semens infrastructure. We serve orsupport all of the different business students
with their needs. And you mentionedwhen I first met you that there's several
other research groups that you guys areconnected with right that are serving different departments
(03:43):
as well. Can you remind mehow many other groups there and kind of
some of the other locations you guysare spread across. Yeah, So our
main hoppus in Princeton in New Jersey, and we also are full function internationally,
so the US it's a Princeton,Brooklyn and Charlotte, and then we
are mainly located in Germany, inIndia and China and have a network of
(04:06):
over two thousand people in this technologyworld serving our business units. Wow.
And how often do you communicate withall those different groups? Would you say
you guys are pretty siloed or you'rein constant communication. We in constant communication,
exchange, some projects we do together, project proposals together. So mainly
SEMA Technology is divided into the companyCore Technologies. This is focused investments by
(04:30):
the full company in different research activities, if you will. And my group
is part of the Siemens Digital Twinand Simulations technology, but we also have
groups for robotics, cyber security,and many, many more. There's an
awesome link on our website. Wecan put it in the show notes and
people can learn where they're absolutely andI think our audiences, if they didn't
(04:55):
already know, they're definitely warming upto understanding digital twin better because recently we've
done several interviews, of course justat Stevens Realized Live in Las Vegas,
and even before that, we weretalking with with v Foyer and Brenda Disher
where we got a nice introduction toa lot of the digital twin kind of
technology that's being developed and obviously alreadyout there serving the industry. So excited
(05:21):
to dive into that more. Andbefore we get into the specific projects and
technology, let's go back a littlebit. I'd love to learn about your
previous kind of education. What gotyou involved in this space? You know,
I did a little bit of researchand you know, learned that you
were working on kind of geo informationand research in that space, and in
(05:44):
some of the interviews I stopped alittle bit of your past interviews. I
really liked that you talked about kindof finding that balance between having an impact
but also still getting to do theresearch and be supported in an R and
D kind of space, right,So I'd love to hear how you started
and how you got to kind ofthis role now, happy to I will
(06:06):
try to get it. Short formaleducation in geoinformatics geosciences back in Germany and
I did my PhD in that fieldand we have as something diploma which is
like a master's And it was aninterdisciplinary study, so we learned a lot
about how to integrate between the computerscientists and the geoscientists and back in the
(06:29):
day GPS, GEO information way finding, trablement, sales problems where all things
we looked into, it was reallyfocused on the computer science, so I
learned how to program. And whatreally changed my view of not the world
but what I want to do wasone of the projects I working as a
students tracking mis in a cage realmice and to help the scientists in the
(06:54):
biology the department to understand when andwhere the mice move to have then this
track into their research for Alzheimer's,so to really make information available, make
it accessible, translate data into information. That was a really interesting project for
me. And from there I foundmy first job in a company that does
augmented reality, so superimposed information,digital videos or little animations into the field
(07:20):
of you, so you have kindof Superman's superbum power as the user.
And from there this was really startup, product based project based. Was lot
of programming and the research aspect gotlost in that field for me. And
then through chance, I had theopportunity to start at Siemens in Berkeley for
(07:43):
technology and to go back a littlebit to the research side of things,
have more time to find the bestsolution, to develop something that has more
impact than a little application at onceon your mobile phone. And now I'm
so fortunate to have my own researchgroups so I can expand what I could
bring to the table. They arenow also with semantic and information domain knowledge
(08:07):
and then the machine vision piece.Yeah, that's so awesome to now be
running your own space around these aroundthese concepts. And I have a few
questions going back to some of thoseprojects you mentioned, because they sound really
interesting. So with the mice trackingyou said essentially is contributing to research around
Alzheimer's, could you tell us alittle bit more about how that was connected
and what were some of the likewhat was the hypothesis for the project?
(08:30):
Yes, I mean to take alittle bit of memory, but the main
idea was to put mice in abigger space, and we designed a space
in a way that we could understandwhen and where they move. And the
hypothesis was the more you move,the more enrichment you have s a mouse.
The few are your symptoms in Alzheimer'sor you like Alzheimer's. The mice
didn't really have that disease, buta genetic disposure for that, and we
(08:56):
could then prove or see how muchthe the mouse would move, where the
mouse would go, what the mousewould do. They had r F I
D transponders and antennas. We designedthis in that cage and we could really
then see where they go based onthe RFID text and help the researchers to
really understand how long were they're drinking, how long were they're on the level
(09:16):
where there were more toys and thingsare like yeah, interesting. We had
a project a while back residency aresident at Design Lab who was working on
emotive robots for Alzheimer's therapy, andthe idea was creating essentially like a robotic
pet that by feeding the pet,which would be like charging or something like
(09:39):
that. By having this routine withthe pet, the patient with Alzheimer's would
be reminded to do some of thesethings for themselves. But I think also
speaks to a little bit about whatyou're saying about having a routine, having
basically needing to move around and dothese different tasks can be beneficial. And
I think decreasing or slowing down someof the symptoms, as you mentioned exactly
(10:00):
before you get to US based wayof unfortunately. Yeah, yeah, really
interesting, incredible work that's being donethere. And so another thing you mentioned
you were in AAR. It soundslike rather early on in your career.
What year was that that you werefirst working on that project? Now I
asked you a question. It musthave been sometime around two thousand and twelve
(10:22):
eleven, something like that. Ibelieve I started in Munich and that company,
and with that company I transition thentook to the US and worked on
different projects. And what do youthink. I know, we'll get into
this in more detail, but ifyou could distill you know, one thing
that's been most surprising. You've seenhow the technologies evolved and progressed in that
(10:43):
time since you first started working onAR to now what you're doing in that
space. This is a great questionbecause when I started AAR, it was
a gimmick. It was something ontop of a TV magazine way at the
schedule, and you could watch littletrailers of the movie or a little chemouse
really in entertainment, entertainment gimmick marketing, and maybe we had one cute project,
(11:05):
will be had a reminder or itwas a reward. The AAR was
a reward for kids when they learnedhow to do party for kim. But
it's still important. It's still moreon work. But now also being at
the Semens side of things, wecan really use a AR to support the
human workforce, to make sure thatyou will be well educated as a worker
(11:28):
in a place where you assemble things, and also help for quality or showing.
So this technology really went from anextra nice to have gimmick to something
that can be really essential to keepworkers educated and also to train the workforce.
And this was something I got tosee firsthand in person at the Siemens
lift facility in Detroit recently and alsothe MXD facility in Chicago. So I'd
(11:54):
love for you to walk the audiencethrough since some of them, most of
them are going to be listening,so they have we have to help them
visualize what does that really look likefor the workforce. So you can walk
through one example, maybe in detail, but I don't want to give it
away, so I'll let you takeit away. But kind of how does
that, how's that set up?And what is the employees kind of experience
(12:16):
getting to use this AR overlay.And there are many different flavors, but
one of my favorite projects that wedid is the idea also that you have
workforce being in the workplace i'd sayfor thirty years, and at some point
this person is going to retire,it's going to get sick, but all
the knowledge is in the head ofthis person. It's almost you drive stick
(12:39):
shift and then you are supposed toexplain it to somebody else in your own
words. It's really hard sometimes totranslate emotion into a sentence own words.
So we had a first project totry out how we can use AAR on
that world that we just watch someonethrough a camera system. We train the
underlying vision system to understand what doyou touch? What you grab? I
(13:01):
assemble something, and then the systemcan detect the things you touch a little
light. How are we built upand then put this information blue on top
of red, on type of greeninto an underlying domain knowledge model. This
is on top of this, ontop of that. We could save that
and you can do it with anytype of camera, but we use the
(13:22):
whole lens one of our favorite devicesbecause it's hands free. You word on
your head and you can use yourhands for your normal task. And then
in a set and set up setup, once you have this knowledge recorded,
you give it to a nervous workeror anybody else you want to train,
and you can see these instructions youhave like an Ikea or Lego instructions
in three D right in front ofyou. You can almost touch the instructions
(13:45):
and you just follow those and againa vision algorithm can capture this, play
it back to make sure that youhave done the right part, and really
walk you through all these assembly steps. Really mind blowing stuff. When I
saw it in person, who ispretty exciting. I mean, it's like
almost a video game overlay in youknow, over your hands that are right
in front of you, and itis trippy. I still, even though
(14:07):
I've gotten, you know, moreand more experiences to use AR and the
hollow lens and different technologies. It'sstill so crazy to see it happen right
in front of you, like youknow, you're just looking through the glass
and still seeing your physical environment andthe things that are on the table and
whatnot. So I'm curious with that. Well, I think something that's really
awesome about what you're described as faras training the model by just having someone
(14:31):
do the task, the expert orthe pro that's been doing it forever,
do it how they've always done it, and there's not this additional layer of
needing to which I think is probablya barrier to entry when some businesses think
about adopting this kind of technology.You don't need to make a three D
model of this and then animate thethree D model to do it, and
(14:52):
then that's what's going to be programmingthe instructions in the future. It's literally
like you put on the goggles andyou've going to I imagine, you know,
you go into the setting where you'retraining it, and you're setting the
standard essentially, and sometimes the mosteasy solution is the best. As they
mentioned, we don't need to fancythree D models and shades and textures and
lights. If I just want toshow you what to grab it can sometimes
(15:15):
be just as easy as an errorand a description and the text grab the
microphone and this is will be reallysimple. And this also takes off the
extra cognitive load to now translate whatdo I see? This is? What
do I do? Now? Ijust point with an error on the object
to grab it, and that makesit quite easy for people to follow instructions
exactly. And it actually reminds meof another project we were involved in,
(15:37):
the Hackaday Prize winner for I'm totallyforgetting which year or was. They're all
a blur now, but I believeit was twenty eighteen. Was the Dexter
robotic arm. And what was somethingthat was I think a big kind of
added value that you know, theyreally promoted how they differentiated from some of
(15:58):
the other arms that wend the marketat the time, and they were really
kind of targeting when it came tothey did a lot of education, but
when it came to kind of businessapplications, they were targeting smaller to medium
sized businesses that couldn't afford, youknow, a ten thousand dollars robotic arm,
but might be able to expedite afew processes with a smaller arm that
was easy to program, and theway that they made it quote I'm doing
(16:19):
air quotes the way they made it, you know, quote unquote easy to
program. As they had this followme mode which was great where they could,
you know, physically someone you putit in the follow me mode,
the robot arm, and then youcould move it around, you know,
just whatever speed is comfortable for youwith your hand. You move around,
you know, the head with whateverfixture it had, if it's a grabber
(16:40):
or whatever, move it, pickup the thing, move it to the
next area, drop the thing,you know, whatever the step might be.
And then in that way it wasa lot more approachable for folks that
might have a business where they're noteven involved in technology or programming. They
don't want the overhead or the workto you know, take that on and
it could be a lot more approachableand not intimidating. And of course they
(17:03):
also had a great support network whereif something went wrong they could help,
which is important. But that wholebeing able to program these technologies without program
exactly without programming in an intuitive waywhere you're really getting to just use what
you know, is really powerful.Absolutely agree, and especially when we when
we see we wanted this level ofentry, keep it low. We want
(17:27):
to make this accessible for us manypeople in the workforce. We see a
lot of people quit change jobs.And if we have these easier ways to
interact with the things we need whileassembling something, pick and placing, this
will help really to keep the workforcein the workforce, and we don't have
to go all the way out we'llreally complicated programming, which will not bring
(17:48):
a benefit to the solution of theproblem. Absolutely. And so what's something
that you're really excited that you're currentlyworking on. It can be related to
this or another project you want totalk about. I'm really excited that we
now can take some of the thingswe learned from the digital twinning to a
different field. Siemens we make trains, as I mentioned earlier, we make
(18:11):
industry equipment, but we have aproject right now that we really love.
We build a digital twin for acoral reef. So we are able now
to take a lot of our learningsand bring this to the world of sustainability.
And this really changes also for usagain the gears before. As I
mentioned in the very beginning, interstipinarywork science geo science. Now a little
(18:33):
bit back in that world. Wework with the WOHI, the Woodsto Oceanographic
Institution, and it's really great todo something that might have a bigger impact
for our planet. And we arereally excited to build this digital reef,
the twin of a coral reef,and you can already see some videos.
We have a website. We'll alsohave to allow that we'll have a catalogic
(18:55):
link. Yes, this episode definitely, but this is really exciting to see
how we can transition now in adifferent domains, still using technology that we
have researched in a really industrial focus. Absolutely, and what's been the most
kind of fun part of that project, Like I remember you were talking to
me a little bit in our previousconversation about some of the communication, you
(19:18):
know, process understanding, you know, being able to kind of speak the
same language, which was a challengeat first, and you guys kind of
got through that. But working withnew collaborators, collaborators, I'm sure it's
always fun and interesting and new youknow, adds a little spice to the
experience, it adds a lot ofwarning. But it was really great for
us then to see once we couldbreak through to really understand there is a
(19:41):
vision of someone who has touched thecorals that are dying rapidly, and there's
us in front of a computer typingcode and then really bring this to life.
And the first time you could showvisuals to our collaborators and see what
we could do out of data.That was a really great moment to translate
(20:03):
data into inflammation and click for them. Finally, I just had a conversation
with Jay yesterday actually, who usedto work who works at a number of
fabrication facilities, mostly supporting innovative technology. They just were advising on space in
Detroit that's being funded by Ford andNew Lab and so similar to kind of
(20:26):
Design Lab, a lot of differentstakeholders and whatnot, And we were just
talking about kind of our passion forhands on building and the satisfaction of bringing
something from concept to reality and howempowering that can be. So I feel
like it's similar when you're working withcollaborators who are maybe not used to that,
they're focused more on their experience withthe environment on their you know,
(20:49):
I'm not totally sure it was anonprofit right and they're working on I imagine
kind of environmental conservation and those sortof things. Being able to have that
kind of aha moment where it clicksinto place, they see what went from
your kind of napkins get right tohave a realized project. And what you
mentioned. I think that's where thecrooks is because we cannot show the parts
(21:10):
because our parts is code bits andbites. At least showing the wooden mockup
or anything like that. We hadsome drawings and some some mockups, but
it's aesthetic and it looks in acertain way and you can't really test it.
It's all digital. But once youhave something, it moves a little
bit, there's an animation and yousee the different colors, and that comes
(21:30):
to like, that's exactly the AhRmoment. And then also you see,
Okay, they told me I wouldn'tsee anything the first x months because they
just sit there and have to codeit then and we'll be ready. Yeah,
that kind of delayed satisfaction, right, but it's worth it and the
anticipation. But it really work exactlyexactly. And so what's the kind of
next phase of that project? Whathappens now that the Coral Reef Digital Twin
(21:53):
exists. We're still in the middleof making it. So we just finished
almost the first module and there willbe a few more to come, so
there will be different simulations edit on. So first was just to make the
frame, if you will, andto have the different types of information temperature
and the benthic information. So whatactually grows is on the core reef.
(22:18):
All is visible and it's an ongoingproject and I think we have at least
another year to make it a fullyfleshed digital reef to try out and work
with. So once the coral Reefit's public, people can access it online
and it's the idea really for communityengagement, for people to be able to
understand and explore even if they can'tgo there in person, and then be
able to contribute to kind of theconservation efforts. It's many faults, so
(22:41):
the main focus is to build thisfor scientists and then to help scientists as
a knowledge tool, the decision makingtool. So this is the first group
of users where you'll have really heavycompetitions in the back end and then you
all can see different plots and seedifferent maps and then see what will happen
(23:02):
if and what did happen when.That's the focused number one. But in
the end, the big goal anddream is to have someone what you said,
being more on the educational space,show people what happens around our worlds
because we can't all go to Palmyra. Is the act all that we're working
on right now, so it willbe more tangible and we can see that
there is an impact and there isa correlation between what we do here and
(23:26):
how it impacts the environment. Alsohow it will come back to ask because
the whole global ecosystem is really wellconnected and I think a lot of us
feel it already with a drought herein California, wildfires. Now we had
this in New York as well,so we really have to watch out for
our planet. Definitely, we'll haveto see if there's a way we can
do some sort of workshop around itat Design Libe in the future, because
(23:48):
I think it'd be really interesting.Obviously, when I ask you the question,
my mind goes to, like communityengagement. Are we going to get
people learning about it? But obviouslyI understand that the scientific advantages are really
the priority. But it'll be greatonce it's more widely available. We can
see different ways we might be ableto interact with the top future awesome and
(24:08):
so we'll remind a little bit andgo back to the kind of workforce development
tools that you guys are training andworking on. I think I was talking
to you about how many, becauseI think there's a dream of kind of
these factory spaces incorporating this at everylevel, right, like being able to
(24:30):
have the tool as holistically as possible, supporting their their you know, goals
of whatever it is of what they'rebuilding, of their team supporting it.
But obviously there's an adoption rate wecan't really ignore. There's some thresholds of
you know, that people have toget past, and I think also companies,
depending on their budget and all theseother factors, may not be able
(24:52):
to do it all at once,right, But do you have some customers
so far or clients who are integratingit at like the fantasy level, like
where it's really shown to its fullest. I'd say still in its infancy,
and what I describe what we dois mostly still in our laps, and
we just start talking to our havingbusinesses, and we talked to warehouses,
(25:15):
we talk to assembly facilities. Weshould always think about safety, especially if
we work with heavy equipment, andadoption rate is really slow. I think
what some of our clients and alsoother customers are more into its smaller solutions,
more short term solutions and habituals.They can kind of fit into what
they're are and have a service engineerand that needs a little bit short term
(25:38):
help. But what we want tobuild is really the whole package of the
domain, knowledge of the human andthe loop and then also to have this
feedback loop about what we assembled.So there will be a process. And
also, you know, workforces hasa lot of different ages and age classes
and also not only our physical age, but also how long have we been
doing tasks, So we have tobe really adaptive to the needs. A
(26:00):
lot of people still work with pencilon paper, and this will be a
slow process to integrate. To givesomeone again a super fansy head looking device
and now this is filming me andwhat I do. So it's it's a
slow process to bring this workforce.Definitely all of the kind of I think
typical responses around new technology in general, right that so many of us face
(26:25):
in the industry. Absolutely, it'sone point I want to make is that
with this technology, we don't wantto displace human workforce, We want to
enable it. We want to empowerit. And also I mean some task
I think we should do through roboticwork for something which is not safe,
which is really heavy, that's howgreat body unhealthy. Yeah, that's all
(26:47):
great, But a lot of tasksyou will keep the human workforce and we
will have this technology for you toempower you, to enable you to put
the human in the loop. Andthat's what's the fun part. Also about
this part of the reason search.We will not replace a human, but
we will enable and empower a humanworker. Definitely. That's so important to
note and something that came up ina conversation I was having with this colleague
(27:11):
Fernando, who works for Citizen Robotics, which is a three D cement company.
They're working on applying that technology tothe construction industry. And something else
I think on the note about notdisplacing workers but really also adding to the
workforce is you know, he wassaying, if everybody was building houses right
(27:33):
now, we still wouldn't be ableto get to the numbers that we need
to to reach the demand right inthe next however, money years who was
saying a few decades right, Andso I think there's that piece too that
it's almost like even if we hadeveryone we could recruit, you know,
in the workforce doing these things,the need has it's far outweighs you know,
(27:55):
what we're capable of achieving just Ithink with humans at this point,
and with humans that to do thoseroles right, So you do need I
think, the addition of some ofthese technologies to help make processes more efficient,
like you said, replace some ofthe less safe processes or less healthy,
you know, kind of roles,and then be able to also augment
and enable the rest of the workforcewho's contributing to reaching all of the necessary
(28:21):
output. It's absolutely this is whythey has not one solution for all of
our I don't want to say problems, but for the need we have,
and we need to put mix andmatch. Yeah, definitely, And I
think, like you said, that'swhy it's so important to have modular,
adaptive solutions that can kind of fitin where needed and not again replace or
takeover. We're going to take aquick break. When we return, Mariike
(28:44):
shares what's next now that the digitaltwin of the coral reef exists, and
later we'll learn how urban hiking andher dog inspire her to learn new ideas.
Welcome back to the bomb. I'ma genta strongheart, and I'm back
with Semen's research scientists. But Iget Klitzler her work with AAR and creating
(29:08):
digital twins is being used to makea broader positive impact on society as far
as the augmented reality solutions. Isthere anything else that you guys are in
development in this space that you arealso excited to kind of get more attraction
(29:29):
on or be pushing further. Andyou mentioned MXD earlier. This is one
of our also funding bodies for usand we just finished a project where we
did anomaly detection during the assembly work. Now it's not a human necessarily in
the loop. It's for blue dotson solar panels and to make sure that
(29:51):
these dots are all correctly applied sothat the solar panels will be able to
do the task they need to whenthey go out to space. Actually,
and that was a really great startingpoint to look more into anomally detection in
a more automated way. And that'sanother branch and I in the end they
will all be able to relate modularizing. As you mentioned, when do we
(30:11):
need to detect anomalies when do weneed to support a worker. But this
is something that we do a lotas well, visual perception type of tasks.
Is there a favorite kind of usecase that you've seen. You were
telling us about the kind of uniquesituations with the train factories, which was
really interesting to hear. You knowthat you guys were touring, but as
(30:33):
far as places that you've gotten to, you know, meet individuals working in
these spaces, but also get anew perspective at a process you maybe weren't
familiar with before. I mean,as you mentioned that the train manufacture,
I see train manufacturing science because thiswas really eye open because made me think
about the manufacturing of courts. Wesee these videos of big com manufacturers with
(30:55):
conveyor belts on the robots and everythinggoes in big quantities. Everything almost everything
on these shop poses automated. Butthen you see the opposite side where you
have these huge pieces of equipment thatyou have to bring. So we see
there's a huge range even if youmanufacture something that is very very technically,
very thought out, that sometimes youneed a lot of menual labor and that
(31:19):
this was interesting for me to seethem. We have super high tech facilities
also Seemen's own facilities. When youbuild electronics versus something that is a little
bit more rough and tumble, I'dsay, yeah, is there a project
that you can't wait to get yourhands on in the next like five or
ten years, Like would be yourdream project to work on the EC maybe
coming down the pipeline or maybe youdon't need to see it on the rise,
(31:42):
and you're like, I want toget this. I want to get
some technology to take it on thisout there. This is more like I
think the coral reef one is alreadysomething to really brand out, but everything
where we can put technology to gooduse. And I'm thinking about safety applications
where we have a mixture of theAAR and machine vision technologies to keep workers
say that we alert someone before somethinghappens, so a little bit promitive and
(32:07):
prediction. It's like, oh,if you go there and make sure you
wear your gloves if you go whetheryou know a fret fell down the train
last week, so you really canmake sure people are going to be safe
while they will do their work.And that would be really something to bring
all these technologies together for a greatergood. Absolutely and with your lab.
You guys work with a number ofstudents, right they come through the space
(32:30):
as interns I would imagine and supportingthe research that's being done with that.
What do you think will be changingas far as how students can prepare for
opportunities in a lab like this inthe future, if people are interested in
getting involved in this space, maybeyounger listeners who want to prepare to be
in you know, this kind ofinnovative research around new technologies, growing technologies.
(32:52):
What's kind of the pathway you seeor any recommendations you would give for
them. I mean, times definitelyhave changed, and we sue technology much
more ubiquitous around us. Even thoughwe think everything is at our fingertips,
someone has to make it so.I think if you want to be in
a space like this for technology computerscience, these skills will be the key.
(33:14):
You have to know how to program, otherwise you can do I mean
the different roles along the line.You don't necessarily have to be a programmer,
but you have to understand what ittakes to bring to get from A
to B, which is mostly programmingin that space, and ideas sometimes they
come overnight, but sometimes ideas takea while to write they come by a
(33:36):
chance. Just be open minded andtry to experience as many things as you
can so you will be able toconceive these ideas and then also communicate the
ideas. I think communication will bea key to really what this is,
what I want to say, whatis the what the person understood, to
really make sure you can be onthe same level. Definitely, I think
(33:57):
that's like those are good. That'sgood advice for anyone who's looking at,
you know, kind of their futurecareer ideas and things and considering what paths
they want to take. Like theskill sets. Absolutely staying curious and open
and observing and trying a lot ofdifferent things. That's how you learn,
you know, creative problem solving,which is useful across disciplines, across industries,
across rules. And definitely that communicationpiece, I mean that's always huge
(34:22):
because at the end of the day, we are humans working with humans,
and you need to know how tocommunicate in different spaces and effectively to give
big projects like this important language Javaor CCU plus plus they will be away
as well. You just have tothen find the right person that's willing to
translate this into maybe more broader humanunderstandable spoken words. Definitely, all right.
(34:45):
So I want to be conscientious ofour time here, and I'm going
to wrap it up with just twokind of rapid by our questions that we
ask all of our all of ourguests on the show here. So,
what's one thing outside of technology that'sinspiring you these days. I'm a big
animal lover and I have a dog, and to learn how to communicate with
(35:06):
him as something really interesting to me. To have the science behind the human
and the animal behavior and learn everythingabout positive reinforcement training where it can really
injured out into that science. Yes, do you have your dog? Have
you gotten him the giant buttons onthe floor that he presses to say I
love you and I want food?You know? He tells me he loves
(35:27):
me by licking my face. It'sfunny. I actually I can't remember the
name of the book. I'll haveto find it and tell you after.
But I was reading this book recentlythat was about this chemist. It was
fiction, but I think there's sometruth to this because I looked it up
afterwards. There was this chemist whowas trying to train her dog, and
like they joke throughout the book,you hear the dogs perspective as well a
(35:52):
few times, and he says thatby the end, you know, he
learns like three hundred and fifty words. She's like really adamant that she can
teach the dog all these words.Everyone thinks she's crazy, but I think
this is actually a possible thing youcan do if you're really diligent about it.
Because if they can understand walk andyou know, food or whatever it
is, why can't they learn more? And it's the semantics and NaNs.
(36:13):
I can't say crocodile and means something, but it's the way you teach and
the repetitions. It's really interesting.Yeah, So I hope that you're teaching
your dog for one hundred and fiftywords. I'll be finding out following up
in the future. I want tomeet this genius dog. What's your dog's
name? Copper? Copper? Oh, that's so sweet. I like that.
That's a good one, pretty original. I don't know if I've heard
Copper before. And also relevant tothe electronic space, I actually want to
(36:36):
if we have some other ees thathas some you know, pets named Copper
and things all right, And lastbut not least, what is on your
personal bill? Of materials, Well, my bill of materials is more a
bill of events, of things likenon tangible things. Activity is exactly.
It's also a pro tip since COVIDurban hiking. It's just back in the
(37:00):
day you say you're going to walkit, but especially here, it's urban
hiking the hills. It's the hillsin San Francisco, and you discover a
lot of great things. Go ona walk, put on comfortable shoes or
Gobert foot maybe not here, andyou will discover a lot of interesting things
people, pets, animals, coyotes, skunks, and a lot open your
(37:23):
mind and really look for the smallthings. You will see little plants just
growing in the crack off a littleplanter, and that's really urban walking,
urban hiking. And then go tothe beach another thing if you want to
relax and just see what nature doeswhen the waves come in and the windy
and grade day. It's also alwaysgreat and just kindness. Let me up,
(37:45):
kind to one another and our petsand our co workers. I think
that will be will be something thatmakes me thrive, and I would like
to share that's beautiful. I thinkthose are super important and relatable to a
lot of folks outdoor, time andkindness the key ingredients to a good life.
So thank you so much, andI think we'll definitely follow up this
(38:06):
conversation in the future. I'm goingto find out about how we can incorporate
the Digital Coral Reef at Design Labfor some upcoming festivities. And yeah,
thank you for the time today.Awesome, Thank you very much. That
was my discussion with Maricha Krizla,who heads up the Smart Machine Vision and
(38:30):
Visualization Research Group for SEMENS in Berkeley, California. I think it's so awesome
to learn more about some of thepractical workforce training applications for r and I
can't wait to see how digital twinscannot only help environmental scientists with their research,
but also engage the broader community andallow them more engaging access to that
research. If you like The Bomb, don't forget to subscribe, rate and
(38:57):
share the show wherever you get yourpodcasts. You can follow supply Frame and
Hackaday on Instagram, Twitter, LinkedInYouTube, and Design Lab at supply Frame
Design Lab on Instagram and Twitter.The Bomb is a supply Frame podcast produced
by me Magenta Strongheart and Ryan Tillotson. Written by Maggie Bowls and edited by
Daniel Ferrara. Theme music is byAnna Hagman. Show art by Thomas Schneider.
Special thanks to Giovanni Selina's, BruceDomingue's, Thomas Woodward, Jin Kumar,
(39:22):
Jordan Clark, the entire Supply Frameteam, and you are wonderful listeners.
I'm your host, Magenta Strongheart.See you next week.