SmartAP 3.0 Pro Autopilot is the high-level flight control system for multirotor Unmanned Aerial Vehicles with the capability of fully autonomous flight including take off, waypoints flight and landing. The hardware core is powered by STM32F4 microcontroller from STMicroelectronics, 9-axis Inertial Measurement Unit and the latest GPS/GLONASS receiver from UBlox providing outstanding flight precision. SmartAP has its own Configuration and Mission Planning software called SmartAP GCS. The system supports any type of multirotor UAV with the highest navigation and control accuracy. SmartAP Autopilot was developed using the latest electronic components available at the moment, which allowed to achieve such amazing results!

General features of the system:
- All-integrated system approach - plug and fly
- Extremely stable flight in
- stabilize (user control)
- position hold (semi-autonomous control)
- auto (fully autonomos navigation and control)
- MAVLink compatible Ground Control Station protocol
- Accurate Position hold
- Horizontal: up to 40cm
- Vertical: up to 10cm
- Manual altitude override
- Fully autonomous waypoints flight
- Autonomous take off
- Autonomous landing
- Return to Home mode
- Low battery detection and Failsafe triggering
- And many many more...

The system installed on 960-sized hexacopter:

Hardware:
- Powerful microcontroller 32 bit 168 MHz STM32F4 ARM Cortex M4
- Compact board size of 8x8 cm (3.15"x3.15"), weight 60g, 6 layers PCB design
- Power supply from main LiPO battery (3S - 8S) support, up to 36 Volts
- Power supply from BEC 5V support
- 12V, 5V, 3.3V generated onboard
- Integrated GPS receiver (UBlox NEO8, GPS/GLONASS, up to 24 sats, 10 Hz) active antenna, exnternal module is supported as well
- Integrated telemetry module (100 mW), external module is supported as well (up to 5V@1A powered)
- Up to 24 PWM I/O support (5V out, high-power), SBUS support
- USB interface for configuration / firmware update
- 6-pin JTAG port for programming / debugging
- Various communication lines (UART/USART, RS232, I2C, SPI)
- MicroSD card driven by 4-bit SDIO interface for data-logging / parameters
- Backup battery for real-time clock and GPS receiver
- Integrated main LiPo battery voltage monitoring
- 4 ADC inputs, battery voltage / current monitoring
- Electromagnetic sound audio indicator
- 4-channels bright LED support (up to 100mA/ch)

The components of the system:

SmartAP Ground Control Station and Configuration Tool
SmartAP GCS (Ground Control Station) is the software application which allows you to plan autonomous mission for your SmartAP Autopilot as well as control the UAV using intuitive high-level commands.
General features are:
- Flexible and user-friendly interface
- One-click commands for high-level control
- ARM, Take off, Loiter, Land etc.
- Status information display
- Real-time video feed and overlay
- with video capture device
- Major flight information display
- Mission planning using waypoints
- Customizable waypoints' settings
- Flight logging support
- Realtime plot of all parameters
- Integrated log analyzer
- Points of interest selection and control
- Terminal window

Ground Control Station Mainwindow

Supported Airframes
- Quadcopter Plus 4
- Quadcopter X 4
- Quadcopter W 4
- Hexacopter Plus 6
- Hexacopter X 6
- Octocopter Plus 8
- Octocopter X 8
- Tricopter I 6
- Tricopter Y 6
- Quadcopter X 8
- Hexacopter X 12

Ground Control Station demo video

User's review video:


More information and ordering at: http://sky-drones.com/

Kirill
Sky-Drones.com

Views: 5925

Comment by Andreas Gazis on January 28, 2016 at 12:26am

@ Darius. To the best of my knowledge, PX4 was developed at ETH Zurich but manufactured and sold to the wider world via 3DR (Pixhawk is effectively PX4 v2). The original software stack (from ETH) is the PX4 stack but then the APM code was ported to it, so PX4/Pixhawks can run either PX4 or APM stacks.

@Marc. To begin with, although it might be hard to go huge (DJI size), the story of the Mexican boy who built upon the open Arduino platform (at the time when the hobbyists choice was extremely limited), moved to the US and founded a company that became a poster boy for the media is not exactly bad. Even if 3DR vanished tomorrow, Jordi and Chris have every reason to feel extremely pleased with themselves for what they have already achieved.

As for ready to fly out of the box systems, 3DR came very late to the party and with an underwhelming gift. 3DR started as an autopilot company and as such, it was vulnerable to the likes of Hobbyking and RCtimer. When finally they came up with Iris, the market was already flooded by DJIs, Parrots and the like. Unlike other manufacturers who went for integration from the start, Iris did not seem to be as polished (not talking from experience, did not own an Iris, just forum chatter). My guess for the reason is 3DR's inertia given its original geeky customer base. The jury is still out on the Solo but pinpointing the reasons why it might be underwhelming when joining so late in the game is hard. Integrated products is a whole different ball game from components and penetrating an established market harder still (why did Android (kinda open sourcey) penetrate very successfully into the established iPhone (proprietary) market? Why didn't Sailfish or Ubuntu or Firefox phones? Many factors to consider and arguing for any of them is harder).

Now, regarding ready to fly out of the box solutions, opening up the code has nothing to do with how polished a product is. Once you offer a very well polished, well thought out ensemble whose parts work perfectly with each other, your product is more than the sum of its parts (if anything, us hobbyists understand much more how subtle a job integration is to get just right) so it's harder for others to eat your pie. A perfect example is Red Hat. Their customers have in-house experts of their own, they can go for purely free solutions such as CentOS or Debian, yet when a business selects an enterprise solution, they buy more than a bunch of code, they get the integration, support and maintenance that their in house geeks don't have to do so they can worry about other things (effectively, outsourcing and centralising some core but boring parts of the IT dept). Red Hat might not be Microsoft but I wouldn't call them a failure by any stretch of the imagination (I would call Sun a failure, which went under although they had some very cool proprietary products).

In this case, if the intended customers are businesses who want something to make their headaches go away and don't mind paying extra for it, opening up the code doesn't matter because what you are selling is not code or components but again, hardware, integration and support, a well polished, finished package. A business customer will choose you because your stuff works out of the box and if it doesn't they can call you and shout at you and it (ideally) gets fixed fast. Opening up the code does not make the integration job any harder, what it does, if it is any good, is invite developers of potentially very high caliber, to contribute to it. When you are just about to start, this might arguably be a good idea (it certainly worked for 3DR, Arduino, Raspberry Pi), you can always change or jump ship, or split into other companies or whatever when you get fat.

Comment by Kirill on January 28, 2016 at 1:32am

Martin, yes, agree with you. It's hard to understand that it's possible to do in a small group of people. However, this is advantage in some way because each developer know the code perfectly and understands the entire architecture of the system instead of adding parts of the code which don't really well fit each other. 

Comment by Fnoop on January 28, 2016 at 1:45am

@Kirill This looks like a really interesting product and does have it's place.  I've recently been looking at flying thermal sensors and even with the cheap ones ($1500-2000) my first thought was worry about the pixhawk/apm:copter reliability.  Not because it's not great hardware/firmware combination or because of the capabilities of the excellent contributers, but because it's essentially not a commercially supported product - despite the fact I've had hundreds of successful flights on apm and I love the hardware/firmware/software.  I think 3dr missed an opportunity by not cutting commercial/production releases periodically and commercially supporting those.  Professional film makers/researchers/industrial filers want a stable, reliable platform to fly on, not a hobbyist codebase that rarely stays still because the developers are (quite rightly) focusing on cool new innovations and technology.  The Redhat analogy is a good one - Redhat essentially takes snapshots of a wide variety of open source software at very long intervals (years between releases) and then supports those snapshots as a stable, commercially supported software.  And this works very, very well - they have a near monopoly on large critical computing environments and they make enough money to support a lot of opensource developers and contribute significantly despite writing very little actual code themselves.  If your product is a professional level hardware with proper QC and proven track record and commercial support then the price of it is excellent.  Shame it's not open source or run a supported version of APM, but if it supports mavlink then it's open enough for 99% of professional level customers.  My only comment is that it looks a bit big?  I personally fly 450/650 size copters and I think it looks a bit big for these platforms?

@Darius you could have done some research and learning before flooding the forums with bizarre statements and questions.  3dr and ETH collaborated to jointly develop the pixhawk and 3dr was the initial commercial manufacturer/supplier - your comments on the 'pirates' thread were really off the mark, as Craig/3dr were joint designers of the original hardware platform.  px4 name referenced the pre-pixhawk hardware separated into fmu and io modules, but now refers to the separate software/firmware stack that ETH maintain for the pixhawk hardware.  The APM software/firmware uses elements of px4 codebase and builds on it significantly for a more user-friendly experience (not a very technical description but the gist of it).  Please stop hijacking every thread with your notions about vibrating tables and magic crystals and your 'fly away syndromes' and hawking your services to everyone who posts a problem.

@Tevada that ebay hardware is brilliant for hobbyists such as myself but it's essentially toy/hobby level hardware, produced by a random chinese factory with zero support and probably poorer quality of hardware manufacturing and QC.  This is no good for professionals which is more where this product is aimed towards.

Comment by Greg Nuspel on January 28, 2016 at 5:07am

I think the biggest problem any flight controller manufacture faces is that they don't have control over what it is integrated into. One trip to a local RC flying field will show that to you. While one person builds very professionally the next will mount parts with bubble gum. So when there is a problem is it really the flight controller?

With a more expensive flight controller you will most likely eliminate the bubble gum builders. This alone will make the flight controller appear to be more reliable.

Comment by Thomas Butler on January 28, 2016 at 5:13am

Like like a great flight controller. Stability looks excellent. Very nice integration! It looks like the controller will accept high level commands via MavLink. It that correct? Is there a document of the MavLink supported commands come with the board? If that is truly a four layer board, perhaps you might consider making it more compact and at least get to down to 50x50mm; just a thought!

Comment by Marc Dornan on January 28, 2016 at 7:27am
@Andreas. A bit of a tangent but well expressed. Thx.

@Kirill. Well at going on 80 comments, mostly positive, you clearly have got people's attention. Look forward to seeing more third party affirmation. If I was not overwhelmed now I would buy one. If I end up flying expensive FLIRs I will definitely revisit.

@Fnoop Dog. I have been thinking and expressing what you are said for a while. 3DR really made a bad, and I think obviously bad, call to throw all eggs in Solo. A pro grade stable open source FC would have been so much more sensible IMHO. It would not have made anyone rich overnight. Now they are in a very tight spot. Did you see that the Dev version of the Snapdragon Powered FC has an Arducopter wiki page! It cost $800 plus, you have to build your own binaries as the libraries cannot be redistributed, and you need to figure PPM yourself. This is does not look so appealing, at least now.
Comment by QuadrotorThai Tevada on January 28, 2016 at 8:54am

Fnoop Dogg  Around to world use DJI so much.

Comment

You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service