Do you want to develop skills to prototype embedded products using state-of-the-art technologies? In this course you will build a hardware and software development environment to guide your journey through the Internet of Things specialization courses. We will use the DragonBoard™ 410c single board computer (SBC). This is the first in a series of courses where you will learn both the theory and get the hands-on development practice needed to prototype Internet of Things products. This course is suitable for a broad range of learners. This course is for you if: - You want to learn how to use learn how to use Linux for embedded purposes. - You want to pivot your career towards the design and development of Internet of Things enabled products - You are an entrepreneur, innovator or member of a DIY community Learning Goals: After completing this course, you will be able to: 1) Know where you can find resources and help in the 96Boards ecosystem. 2) Describe the DragonBoard™ 410c peripherals, I/O expansion capabilities, Compute (CPU and Graphics) capabilities, and Connectivity capabilities. 3) Understand how to navigate and make use of the Linux terminal. 4) Configure at least one integrated development environment (IDE) for developing software. 5) Make use of Git and GitHub for version control purposes. 6) Create and build projects that interface with sensors and actuators through GPIO and Arduino.
Vertical Markets and Path to product
Loading...
Del curso dictado por University of California San Diego
Internet of Things V2: DragonBoard™ bring up and community ecosystem
3 calificaciones
De la lección
DragonBoard Bringup and Ecosystem
Conoce a los instructores
Ganz ChockalingamPrincipal Engineer
Qualcomm Institute of Calit2, UC, San Diego
Harinath GarudadriAssociate Research Scientist
Qualcomm Institute of Calit2, UC, San Diego
So far in this course we have talked about the 96Boards ecosystem,
we have talked about DragonBoard and how the ecosystem enables a wide variety of users.
It enables not just your hobbyists,
but it also enables industry professionals.
If you want to work on a product,
this is a good platform that enables you to
visualize and not only to visualize but to commercialize your product.
So in this section we'll talk a little bit about verticals,
what verticals you might be working with,
and how the ecosystem and the 96Boards DragonBoard 410c enables a real part to product.
So you might be working on various verticals.
You might be hobbyist,
you might be working from home,
or you might be an industry professional working in one of these verticals.
You might be working with robots, drones, compute,
automation, industry automation, you might be
working with automotive environment and so on.
So the verticals keep going on and on and on.
You might come across something that is very specific to your vertical.
And we are going to take a few moments talking about
how not only the 96Boards' ecosystem but also the DragonBoard
410c enables you to kind of commercialize your product,
not just have a proof of concept in hand.
I want to touch real quick on one
vertical that's not really mentioned here but is just as
important at least in
the sense that maybe it's not even considered a vertical but the maker space.
And so makers aren't really listed here,
but makers make up a huge part of the community.
And as a maker if you ever come across something that you're developing,
something that you say, "Oh, you know what?
I'm just going to go do this myself,
DIY," you're one of those DIY makers,
but in the process you say,
"Wow, this is a great idea.
I want to find a way to bring this to market.
What vertical did you-."
This is where you kind of start bringing
this whole video and why we're talking about this.
You are the maker and you figured something out that's really cool.
What vertical are you tackling?
And then how can,
what we're going to talk about in a second,
Qualcomm or Arrow or any of these big industry partners or even yourself,
how can you bring that to market,
so how can you turn it into a product and bring it to market?
Yeah, that goes into our next slide.
So basically when we talk about the DragonBoard 410c,
that has a Snapdragon 410e processor,
and we touched a little bit on this in the earlier videos in the course.
But essentially 410c when we talk about DragonBoard 410c c stands for community effort,
and what that means is Qualcomm with
the Snapdragon 410e part is working very closely with Arrow,
with all the component manufacturers of
all the components that go on the DragonBoard 410c.
They came together, collaborated,
and contributed towards this awesome community platform
that we are talking about right now.
And as we talked about,
so difference between 410c and 410e,
so it has the APQ8016 application processor from Qualcomm and e stands for extended life.
So when we talk about extended life,
depending on the vertical that you are involved with,
product lifecycle really matters.
If you have a part which is only supported for a couple of years,
you might not be able to end up creating a product that has a good lifecycle.
In a lot of different verticals, for example healthcare,
just a certification process is so long that if you have a part on
your product which is not supported for an extended lifecycle,
you're not going to have a reliable product because just the fact that you have to get it
certified for a whole bunch of compliance for that vertical,
you might end up missing the lifecycle.
Yeah, I mean going into something complex like
medical but also going into something simple like
maybe a gateway to a group of trash cans or lightbulbs in a smart city.
I mean you want your product running the Snapdragon 410e chipset to be
stationary in an industrial setting for a long period
of time without the need of tampering or messing with it of any sort.
So I mean the fact that it's supported for such a long time allows people
to develop around these environments and cater to this big IoT boom that's coming up.
So when we say such a long time,
it's actually 10 years, so it's supported to 2025.
Yeah, 10 years. That's a long time for it.
So it got launched 2015 and it's going to be supported till 2025,
and as part of the other ongoing support for
IoT space and extended longevity on different parts,
we also have some other products
which Qualcomm is going to
identify for extended life and 410 is just one piece of the puzzle.
Yes.
So when we talked a little bit about verticals,
you could be doing a whole lot of different,
you could belong to any of those verticals that we had in one of the previous slides.
We are just going to take one of those verticals and break it down to
just kind of visualize how fragmented the vertical space gets.
So if you just talk about industrial IoT,
you can be even under industrial IoT umbrella.
You can have industrial gateways,
you can have home automations,
energy and metering, you can have asset tracking, so forth and so on.
So it gets very,
very fragmented when you're trying to build a product and
it just the process of identifying a part
that is geared towards that particular vertical can be a good Google search for somebody.
Rajan, I'm trying to think of what you would call a vertical inside of a vertical.
It's like vertical squared.
So I mean, yes, like industrial IOT and then you have
all these different spaces that you can work in. So yes, sorry.
Yes. So how does Snapdragon
or a DragonBoard kind of support all of these different verticals?
When we talk about these verticals and then verticals within verticals,
how does it support?
In the end, any vertical you are working with,
you basically have certain requirements for your product.
You need to have certain compute power,
you need to have a certain amount of connectivity,
whether that's Wi-Fi connectivity or
cellular connectivity or some other low range RF connectivity,
you would need to have integrated processing engines like a CPU,
or you might need to have multimedia support on the product,
and that's where something like a DragonBoard
410C and a Snapdragon platform comes into picture,
where it can support a lot of different requirements like those.
Great support to those requirements,
you can basically dig the same platform and use it for all of these different verticals.
But then also, are you going to touch on the kind of the ecosystem around it.
Because I mean, you're saying,
that the Snapdragon in itself is the development platform.
But now you get situations which I mean,
in the theme of this lesson is the mezzanines, right?
And for instance, let's just call out right here.
I don't know, D3 Engineering's camera mezzanine, right?
So you're developing on a vertical that requires some sort of use of the camera, right?
Well now, you want to use the Snapdragon 410E processor,
that chipset, but you also used the D3 mezzanine in your development process.
So what happens after that?
When we go through the development process we will touch on that a little bit,
but like you mentioned, the ecosystem really enables the developers.
When we're talking about being able to kind of cut down on those development cycles,
I think we have a point on the slide which is
basically rapid developing cycles and optimized technologies.
Building a product which you can do a proof of concept
and take it to market with
a short development cycle is really important for some of these products,
especially in this fast-paced technology environment.
Every year products are being upgraded,
every year new technology comes out.
With that said, the ecosystem really does help,
kind of, not only cut down the development cycle,
but also make a very scalable product.
In terms of the development cycle and being able to take something like
a DragonBoard from a development platform
to an actual product which can be commercialized,
it really enables a developers and be,
it walks them through the entire process.
In terms of the actual development cycle rate,
you have the development platform.
You start working with the development platform,
you start working on your concept,
even if you're a hobbyist or working out of your home,
you have a good concept, you think it's
a million dollar concept and you want to start building it,
so DragonBoard is a very good place to start.
Once you have that proof of concept done on the DragonBoard,
using the mezzanine ecosystem,
you could be targeting a wide variety of verticals,
and you could identify a mezzanine that is catered towards that vertical,
or you could end up creating your own mezzanine for that particular vertical.
Once you have that proof of concept with the DragonBoard and the mezzanine ecosystem,
you would really want to do is take that and do a prototype off of it.
Now, you can consider the DragonBoard and the mezzanine board as a prototype,
but what you would really want to do is do your own design.
With that said, with the Snapdragon platform,
what that enables is there's a lot of vendors available which offer the same platform,
the Snapdragon 410E as a SOMs and SBC.
DragonBoard 410C, like we've talked about,
it's a 96Boards ecosystem board,
but it has a lot of different components like the high speed and low speed connectors.
So when you build your proof of concept on a DragonBoard,
you might have to do something with computer vision,
you might end up using the D3 camera mezzanine,
and so forth and so on.
But you might not need some of the components.
You might not need USB connectors,
you might not need GPS on it.
Yes, I want to dive in a little deeper on that because I've talked to several people,
and it is literally kind of one of the biggest points I think,
that should be taken away of your path to product.
And it is essentially saying that, "Yes, okay.
You build a prototype.
Maybe you get 10 or 20 or 50 DragonBoards,
and you practiced, and you build out your prototype,
and you tested in your own closed world."
But by the time you really want to
execute this and you're building thousands of these things,
tens of thousands of these things,
one USB port is going to put a dent in your budget.
You know, I mean, overall,
if you have a hundred thousand units,
and each USB port cost you ten cents,
then you want to knock that off.
So the derivative design process comes into play
big time because if you can save a hundred thousand dollars,
two hundred thousand dollars, a million dollars, then that's resources.
That pays for a resource. That pays for a resource for a year.
Yes. Engineer or someone that you can hire.
I mean, those are the points that you need to
kind of weigh when you're moving from your initial prototypes,
your development phase, which is the evaluate and develop into your prototype phase,
and then on into your commercialization.
You want to knock off all these unnecessary things that are currently on the DragonBoard.
And that's where the ecosystem really plays a part.
There's a lot of vendors out there that have worked with the SnapDragon platform that are
willing to kind of take
up these new ideas and new concepts that the community is creating,
and build on top of that.
We have a quick link right there.
We would encourage you guys to go take a look at that link,
it's for the Snapdragon 410E landing page on the developer network.
The link might change in future,
but basically go to developer.qualcomm.com,
look for the DragonBoard page,
and click on the Snapdragon 410E part.
And there you will see a lot of different options which would walk a developer or a maker
through the process of creating
your own prototype using
the SOMs that are available or a single board computer that's available,
and how to commercialize it.
When you talk about commercialization,
you have to go through some testing and certification,
you have to put support into place,
and all of those different factors that will,
in the end, contribute towards a good user experience for the end product.
There's a lot of people out there to help you too.
I mean, you know, I would go as far to say as,
you know, folks from Qualcomm, folks from Arrow,
folks from these different distributors and manufacturers,
they all want to be a part of this.
You know, make sure you utilize your resources.
Time to pause and reflect on the resources, right?
I know 96Boards also has an initiative
for different verticals along the lines of what we talked about.
Yes, so I can kind of touch on that real quickly.
You can see the link up there,
it's on the resources page.
Again, like Rajan says, pause, reflect,
go explore those Snapdragon 410E pages,
or the DragonBoard 410C page on Qualcomm developer network.
Now, if you want to see some real life uses
of kind of verticals that the 96Boards team has tried to tackle,
as well as some other people in the community,
you can go to this GitHub repository, 96Boards, or github.com/96boards-projects.
It's an entire org actually,
and inside the org you have different repositories that have been geared towards
setting up foundational use cases around a variety of vertical markets.
Now, what's the point of us doing this?
Well, the point in general is to just show people that
these baseline verticals can be
approached and used for platforms like the DragonBoard 410C.
Now, if you consume this,
and you build on top of it,
and you make something bigger, better,
well then I mean, that's your first step to path the product.
I mean, you consume the DragonBoard,
you consume a mezzanine, a particular functionality,
a market that you want to address,
the vertical market that you want to address.
You can go take a chunk of code,
a chunk of open source stuff that has already been developed for you,
that's open and ready for you to consume,
and then make it unique,
make it special for your use case.
That's kind of what that projects org is there for.
If you want to contribute it back,
afterwards, that's always welcome as well.
So I mean, again, open source.
Yes. I would also like to add,
in terms of consuming the DragonBoard and the mezzanine ecosystem,
we talked a little bit about the mezzanine ecosystem and how
it's open source and you can start building your hardware on top of that.
But it's important to also mention that,
when we're talking about derivative design,
taking the DragonBoard, something like a DragonBoard,
and moving towards using something like a system on module,
the entire schematic for the DragonBoard is also available on Qualcomm developer network.
So if you go to Qualcomm developer network,
go to the DragonBoard 410C resources page,
you can actually find the entire bill of materials for the DragonBoard,
and the schematics and layouts for
the DragonBoard is available for everybody to download.
You can take that,
you can do your own design,
or you can identify something like a System-on-Module,
which has the same part.
And since you have done your software design on the DragonBoard and you
use the same Snapdragon part just in a different form factor,
just the System-on-Module, you do not have to reinvent your software code.
So support for that is the same support for a SOM.
It's just the reusing it in a different form factor to better suit the end product.
It's very nice when companies do provide those bill of materials because
you can see exactly what you want on your derivative design.
And in a sense,
where you can get it from and who's going to help you build it,
so very, very nicely put.
This concludes, as far as now, this video.
But please, check out the next one.
You are going to have some homework.
So, we look forward to assigning that to you. Thank you.
Thank you.
Explora nuestro catálogo
Inscríbete de manera gratuita y obtén recomendaciones personalizadas, actualizaciones y ofertas.
Coursera brinda acceso universal a la mejor educación del mundo, al asociarse con las mejores universidades y organizaciones, para ofrecer cursos en línea.
© 2018 Coursera Inc. Todos los derechos reservados.
