Blogs

My latest autopilot with new ADXRS453 gyros is ready for testing.

- Two switching power supplies (Traco Power) at 6.5V and 5V for low power dissipation

-  Separated reference voltage with RLC filter for sensors

-  Separated reference voltage with RLC filter for ATXMEGA's ADC reference input

-  Separated power supply with RLC filter for MCU's digital side

- Two MCU's  ATXMEGA128A1 @ 40Mhz (40+40 MIPS), one for AHRS and one for navigation

- ADXRS453 ultra high vibration rejection gyros, digital output (SPI)

- LIS344ALH Accelerometer

- HMC5883L  3D Compass

- MPL3115A2    I2C Precision Altimeter - internal ready calculated altitude !! - saving MCU time for other tasks

- 6 R/C inputs

- 8 PWM outputs (460Hz)

- 2 PWM outputs for servo camera stabilization

- 8 digital outputs (500mA/each) for buzzer, external LED's,payload trigger,etc

- 2 spare analogical inputs @ 12bit

- 4 general purpose I/O's (I2C, UART, PWM)

- battery voltage monitoring

- Serial COM. for GPS

- Serial COM. for XBEE

- Serial COM. @ analogical input for ultrasonic Sonar

- Four LED's for GPS,XBEE,etc.

I will start doing testings with the AP in high vibration environment - by using unbalanced propellers + motors and see how well gyros will perform. In following days, MPL3115A2 will be also delivered. Just waiting to see the altitude precision of  this new IC.

Will keep you posted about testing results.

Cheers,

Sergiu

Hi all,

after a lot of research, programming, setting and tuning, I'm proud to announce that my Proxima II went in flight.

It was already, but without any autopilot, only to set the right controls, trims, CG and wings angle of attack.

Now it is fully autonomous. This is the more complex autopilot equipped plane I ever build.

Here you can see the plane in full action.

In this video there is a demonstration of the controls possibilities.

The pilot has the following controls over the plane :

• throttle
• elevator
• rudder
• ailerons, differential !
• flaps, with those functions : thermal flaps, crow flaps
• autopilot mode
• coming next will be camera pan / tilt and camera mode controls

Possibilities are :

• roll control over two (ailerons) or four surfaces (ailerons and flaps)
• camber control over four surfaces (I could program it over two surfaces too, but actually I don't need it)
• crow flaps, with roll control over two or four surfaces
• fine trimming of both ailerons and flaps for roll or camber

About the plane :

Proxima II, 2.78m, Selig 3021 (Reichard)

MVVS 3.5 / 1200

Castle Creations ICE 50A ESC

Thomas Scherrer RC system

UavDevBoard 4 stabilization / autopilot unit

And finally an amazing board that allowed me to have all those functions mixed with the autopilot.

FpvKiwi Channel Wizard

This mixer is incredibly flexible in programming.

It is fed with PPM stream from RC rx. The UDB is daisy chained to it and feeds back to it again ailerons control for stabilization and autonomous flight.

Here are the schematics of all the connections :

Here you can watch some more pictures of the plane.

Next plans are to mount a video system and use it as an FPV platform, for long time flying using thermals.

This plane is dedicated to Bill Premerlani, as a tribute to him for starting this amazing adventure, for all the improvements he achieved and for letting me the privilege to travel a short distance on this long road with him.

Thank you Bill !

Best regards,

Ric

The New America Foundation had a conference in Washington DC last week called "Tinkering with Tomorrow: Will the DIY Movement Craft the Future?" 

The full conference is recorded in the video above, but check it at at the 30:00 mark, when Tom Kalil, the Associate Director of White House Office of Science and Technology Policy, singles out DIY Drones as a great example of "cyberphysical systems". That's defined as very low-cost hardware and embedded software--a physical device with sensors and actuators, connected to the network. "A huge amount of innovation is going to come out of this", he said. 

NASA and Lockheed Martin have teamed up to create the X-56A, which is designed to use sensors and fast-response control loops to stop wing flutter, which can be a killer in high-aspect ratio wings. From Wired's DangerRoom:

When flutter does occur in the X-56A’s slender wings, the on-board flight computer will manipulate the flight control surfaces on the wings in an attempt to reduce it. Although some fly-by-wire aircraft have been able to reduce the oscillations, the typical solution is to simply increase the strength (and weight) of the structure.

If successful, new designs could lead to ultra lightweight structures and extremely efficient wings for future high altitude drones expanding the capabilities of the eyes in the sky.

The research could also make its way into the civilian world. Because strength, weight and efficiency are interlinked, many futuristic designs promoted by NASA and others rarely go beyond the design concept stage because there is no way to safely use lightweight structures that lack the strength to handle potentially destructive things like flutter.

From IEEE Spectrum

Here's a quadrotor with giant spindly bird legs from the Utah Telerobotics Lab. It's a passive system which uses the weight of the robot to actuate the grippers, so that when the quadrotor lands on something it latches on and stays. To release, the robot just lifts off, and the grippers let go. Birds use a very similar system to keep themselves perched while asleep, and the leg and toe design was inspired directly by songbirds.

[ Utah Telerobotics Lab ]

Here's it flying:

This is a little off-topic for us, but here's a new achievement in aeronautics: a new paper airplane distance record of more than 266 feet. I'm particularily proud of our former University of Berkeley quarterback who had the winning throw. There was a little sticky tape on the plane, which is allowed by the rules, so it's not 100% paper, but the competitors are being gentlemanly about it..

Hizook has a great analysis of VC funding for robotics companies in 2011. There isn't as much of it as you might expect (note that only one is for a UAV company, Helen Greiner's CyPhy Works, which has yet to release a product). Here's his explanation for why:

My hope is that robotics will get more love in the next year(s), but getting VC funding for robotics is a decidedly tough nut to crack.  Robotics companies have large capital requirements for robot hardware, few potential acquirers, and almost no "Google-scale" breakout success stories (ie. IPOs).  I mean, c'mon... one of the best known robotics companies, iRobot, has a market cap of just $700 Million.  This makes robotics a difficult sell to your typical VC firm.  My hope is that this list can give others courage to pursue "swing for the fences" type projects along with a source for robotics-friendly VC firms.

This great talk, given yesterday at TED, is now live. I was here and the crowd was wowed. Check out the lovely math on simplifying the 12-dimensional math to 4 dimensions at around the 5:00 mark, 

    Just finished building an almost production quality heavy lift foldable Y6 frame.  Bare frame without electronics and motor mounts is 327.4g.  All frame parts are cnc machined.  The motor mounts are molded out of light weight fiberglass reinforced plastic (rear mount), and light weight graphite reinforced plastic (front).  

    The idea was to build a strong, lightweight, foldable, heavy lift platform at the bare minimum of cost.  All the components are covered and protected by an ultra lightweight fiberglass body set.  I was using carbon for the prototypes, but later opted for fiberglass to limit interference issues so everything could be enclosed within the body.  The carbon body weighs in at 80g a piece.  The Fiberglass body weighs in right around 45g (90g for both halves).  

    The arms are 16mm carbon fiber cello wrapped tubes.  The arms have almost a 2mm wall thickness, which makes them virtually indestructible.  40-50% of the arms weight can be drilled out for further weight saving without compromising structural integrity in flight.  The sub-frame assembly was cut from 2.5mm peel-ply carbon fiber panel (very strong and rigid, but at half the price of traditional weaved carbon fiber fabric).

    Everything on this air frame is an original custom design, especially the landing gear.  I have built and flown a number of multirotors with varying success.  The quality and flight characteristics have continually gotten better with each consecutive build.  In a previous blog that I posted "Trilo-Byte preproduction prototype"  I tested the effects of using light weight shock absorbent landing gear.  Due to the efficiency of prototype landing gear I decided to incorporate them into this version as well.  Considering the amount of money that goes into an AP rig why do we still use stiff rigid landing gear when landings on even the best systems are all but perfect?  I believe that it makes perfectly logical sense to dampen the landing struts to minimize excessive force on expensive hardware and components.  The landing gears dampening systems can be adjusted with cheap inexpensive o-rings.  

    I am also finishing up on a production quality quad build using many of the same building techniques.  I have also designed a better method for mounting ESCs', which makes them more accessible and easier to replace in the event one fails; also greatly reduces wire clutter.  I will post the quad photos and subsequent flight videos once I have finished fabricating and mounting the landing gear.  

    The premise behind all of my designs are to make the highest quality air frames at the lowest possible price, but most importantly make them easily serviceable with readily changeable and easy to produce parts.  Best of all, I design my frames around the ArduPilot hardware.   I will post videos and more photos soon, along with more specs.  

Feel free to post your comments, questions, suggestions, or criticisms.  All feedback is appreciated.   

I have an old BNB Digital Power recorder so I wanted to have a look at the hovering power requirements of my quad. The weight of the quad with either battery is very similar.

Quad specs:

APM2

KDA20-22L motors from Hobbyking

2 x Super Simple 25-30A ESC's from Hobbyking

2 x Hexfet 25A ESC's from Hobbyking (2 of 4 were DOA)

10 x 4.5 props from Goodluckbuy.com

Batteries: 4S 2200mAh Nanotech 25C

3S 3000mAh Turnigy 25C

X525 frame

What vibration can do to your model (1:1 example)

I've been working towards a realistic simulation of my high-altitude return-to-launch glider project in X-plane and this is the latest result which some here might find interesting.

The simulation is running with X-plane 9, APM-MP 1.1.36 (to keep the TAS into APM - more on that later) and a slightly modified version of ArduPlane 2.28.

To achieve the drop, I drag the aircraft up to the release altitude using the Local Map feature in X-plane and set the speed to zero.  Whilst this isn't a perfect simulation of the intended release due to the flat orientation of the aircraft, for my purposes it is sufficient for now.  Initially I allow the aircraft to accelerate in Manual mode until about 25kt IAS, then I switch to Stablize, manually manipulate the controls to pull out of the dive as rapidly as practicable before proceeding to switch into Automode.  Eventually I need to code this whole process to trigger and execute all automatically in APM.

On this drop, I set the release point to be about 80 miles 'upwind' of the simulated launch point to make a rough allowance for the drift of the balloon on ascent.  I am using the X-plane's real weather simulation (where it gets its weather data from some unknown on-line and quasi-real-time source), so I couldn't be absolutely sure what the headwind was going to be.  This was the third in a series of flights, where the release point was selected from the experience of the earlier flights to give a likely head wind for the whole descent glide stage.

As you can see from the Google Earth image above, most of the travelling is done above the jetstream altitudes and the aircraft arrives at the loiter point still at nearly 45000 feet!

I logged the data using the data output to file feature in X-plane and then created a KML file using this method which uses this on-line file translator.  Whilst logging the data in X-plane, I added more than just lat-long-alt and included some airspeed data so I could figure out a few other things.

The altitude profile shows the entire descent stage lasting a little over two hours:

and, the speed profile is here.  You can see how the ground speed is fluctuating wildly as it tries to hold the loiter position.  I'm not sure if the peak airspeed of nearly 500mph is achievable either because I didn't make a note of the Mach number, but it was probably less than 0.85.  I suspect this could be an issue for airframe integrity...

More work needed here to limit the speed during the dive phase.

Finally, I integrated ground speed to give me an estimate of the distance I could feasibly travel from the release point.  A pretty impressive number, don't you think?  Line of sight all the way too! ;) 

...and before I forget, here's a few details on the TAS problem I mentioned earlier.

The APM-MP-based HIL simulation up until recently has fed APM with true airspeed (TAS) from the X-plane data.  I need this primarily for the PID loop speed scalar to allow stable control at high altitudes.  The only problem is that this isn't what APM gets in real life from a pitot.  The pitot-static system will give you indicated airspeed (IAS), or more correctly, dynamic pressure, which deviates wildly from true airspeed as density changes.

Michael Oborne has changed this in a later version (1.1.42 at least) so that APM now sees a realistic IAS number from X-plane, however this would mean I'd have to implement the IAS-TAS conversion in the APM code.  I must do this at some stage, but for now having this little MP bug is a handy thing! :)  Lazy?  Moi...?

The APM software mod I've done is an embarrasingly simple one (yes, it's true I'm not a software geek!) where I expand the speed scalar clips from the standard 0.5-2.0 to a randomly selected 0.01-5.0.

http://www.suasnews.com/2012/02/12471/double-o-drone-grasp-lab-plays-bond/

Quadrotors designed and built at the University of Pennsylvania perform the James Bond Theme by playing various instruments including the keyboard, drums and maracas, a cymbal, and the debut of an adapted guitar built from a couch frame. The quadrotors play this “couch guitar” by flying over guitar strings stretched across a couch frame; plucking the strings with a stiff wire attached to the base of the quadrotor. A special microphone attached to the frame records the notes made by the “couch guitar”.

These flying quadrotors are completely autonomous, meaning humans are not controlling them; rather they are controlled by a computer programed with instructions to play the instruments.

Penn’s School of Engineering and Applied Science is home to some of the most innovative robotics research on the planet, much of it coming out of the General Robotics, Automation, Sensing and Perception (GRASP) Lab.

This video premiered at the TED2012 Conference in Long Beach, California on February 29, 2012. Deputy Dean for Education and GRASP lab member Vijay Kumar presented some of this groundbreaking work at the TED2012 conference, an international gathering of people and ideas from technology, entertainment, and design.

Now with a loop at the touch of a button and USB port that allows you do grab your HD images.

The AR.Drone is beginning to be a useful platform, not long before every builder has one.

http://www.suasnews.com/2012/02/12433/pre-orders-to-open-march-1st-for-ar-drone-2/

I just discovered CircuitLab, a online electronics simulator. I haven't had time to play much with it yet, but it looks very promising from what I have seen so far. Especially considering that they just launched the site.

So after another $75, all of Marcy 2's brain finally arrived.  To be sure, Marcy 2 has gone through a lot of failed iterations & airframes.  This time, she's a Wifi monocopter.


As bad as it is for business, the cheapest way to get WiFi in a robot is the $5 USB dongle.  It's not as convenient as the Roving Networks products & no-one else does it, but we already reached the same conclusion years ago & implemented a USB dongle in another gadget.

It's amazing how persistent that conclusion has stayed.  The mane problem is these dongles are their 1 off, single manufacturing runs.  You have to write a new driver every time you order another set of dongles.

As for the Wifi dongle,

 It's the RTL8192CU.

It is built on top of the USB pads, so connecting it is hard.  It appears to use 3.3V & 1.5V inside.  A datasheet for the RTL8188CUS would be nice.


Also, it's finally time to use an ARM.  The last ARM attempt died in 2007, when our home made JTAG got it flashed, but it still couldn't toggle a pin.  This time, we tried as hard as possible to find an Arduino reference design, but there's still nothing & the chip they're targeting doesn't have nearly the horsepower we need.

ARM is definitely still a jump in complexity.  It still needs a ton of connections to get going.  It still needs paragraphs of registers to set a pin.  The mane improvements are only 1 power supply being needed & no external clock being needed.

So the camera interface on the STM32F407 uses a double buffer.  Software can massage small bits of data before storing it in a frame buffer.  If the transfer rate was low enough, it could JPEG compress 16 rows at a time.  It's going to take some 64 UDP packets to transfer an uncompressed 320x240 greyscale frame. There's not enough room on the aircraft for DRAM.

GCC CROSS COMPILERS FOR ARM


In the quest for Marcy 2's cross compiler, we found exactly 1 guide which worked: 


http://cu.rious.org/make/compiling-the-arm-cortex-m4-toolchain-yourself/


The gnuarm toolchain we tried 5 years ago hasn't been updated since 2007.  People have moved to commercial tools on Windows.  The only build system for ARM we found was https://github.com/texane/stlink.  https://github.com/texane/stlink/tree/master/example/blink has a Makefile.

Marcy 2's camera is the 8 year old, famous TCM8230MD.  Low volume options in this department are still limited to ancient parts.

Now that the undocumented footprints for the camera & Wifi are available for measuring, we can finally lay it out.

Sparkfun forums told a tale of various attempts to use the TCM8230.  People have driven it with clockspeeds down to 12Mhz.  They don't know if it can go lower.  They get a framerate of 15fps when driving it at 12Mhz, but don't know if it can do full resolution at 15fps or only 1/2 resolution.

They don't know the output data rate with the 12Mhz clock & 320x240 video.  They've used discrete logic to store the output in SRAM.  1 guy used an AL440B to buffer the data.


There's also the TCM8240MD, which has JPEG compression but is 8 years old.  The hardware JPEG compression makes everything smaller.  The thing is no-one's using the JPEG compression.

According to the Sparkfun forum, they're buying it for the JPEG compression but just using it for uncompressed.  Neither do they know how fast the JPEG output needs to be captured.  It could be 25Mhz, just like the uncompressed data.  JPEG cameras are such an exception to the rule of what's sold, they're almost not worth it.

Most important

is the Leo Laporte ball, $15 at Big 5.  Theoretically, it should keep us awake & keep us from hunching over while soldering.  Who knows how many soldering iron drops & xacto knife drops & it'll withstand.  It does transfer whatever is on the floor to your hands & make it hard to fart.

After scouring most of the shops in my area for a cover/box/hat for my 3DR Quad I found the Orthex 140 0,75 liter kitchen bowl.

The fitting is so good it almost brings tears to my eyes.

It sits on the bottom plate just outside the nut on the arm and I could even reuse the nylon screws on the top.

Highly recommended!

Good news: APM 2 preorders have reopened. You can now choose it with the onboard GPS or without, so can use an external GPS or any other supported GPS module (the Ublox GS406/407 can be used with this adapter, and other modules can be used in NMEA mode). 

All the necessary components for this batch of APM 2s will be coming in by the end of March, so pre-orders placed to today will be shipping in 4-6 weeks. After this pre-order period we will be in full APM 2 production, so the product should be constantly in stock by the middle of April.  At that point, we'll probably also offer versions with external magnetometers, and with an option for different connector pins, including right-angle pins.

We also now offer a version with no pins pre-soldered at all, for those who want to configure it to taste themselves. 

You can always follow the 3D Robotics Twitter feed for more regular updates.

On the code front, the current ArduCopter and ArduPlane supports APM 2, but has not yet been optimized for its sensors. Optimized code will be out in the next couple weeks, and further enhancements will be coming in the months to come. There's a lot of power in APM 2's MPU-6000 sensors, and you'll find that each software update taps it more. 

The first code enhancement, coming in the next week or two, will address the "lean" issue that some people are experiencing on multicopters. This is due to vibration. With the current code, the APM 2 board is more susceptible to this for two reasons: first, because it's a lighter board than APM 1 and as a result has less mechanical damping even when mounted on vibration-insulation, and second because we have not yet implemented software filtering of accelerometer noise.

The reason for that second issue is that in the current code, most software filtering is done in the ADC library. But because  APM 2 uses digital sensors, unlike the analog ones on APM 1 (which go through an ADC), it doesn't use the ADC library. This week we're going to move the filtering to the main ArduCopter code, which is shared by both APM 1 and APM 2, and we'll be able to optimize the settings for APM 2. This should resolve most of the "leaning" issues with a simple software update, which should be out by mid-March. 

There are a bunch of other improvements we're doing to improve filtering, including new test processes and easier ways to adjust the filters, both for the dev team and end-users. We're also preparing to use the MPU-6000's internal DMP processor, as an option to replace the software DCM sensor fusion algorithm we now use, but are still working with Invensense on a legal path to distribute the source code, in keeping with our commitment to open source. Look for that update to be out in a few months. 

The rugged Arducopter 3DR frame is now available in a Hexa version capable of supporting larger payloads and longer flight times. With the stock 850kV (purple) motors, a basic Hexa with ESCs, APM, GPS and an RC Receiver weighs in at 1356.3g and can lift more than its own weight.

This version supports a fancy new feature allowing easy access to the power distribution board. The entire stackup can be removed with thumb nuts to allow access to the central area between the base plates.

Lifting tests were performed with the following results:

Frame Weight: 1356.3g

Battery Weight (4S 5000mAh): 534.0g

Misc. Payload: 533.7g

All-Up Weight: 2424g

Flight Time in Hover: 12:10

Over 24 minutes endurance is possible in this configuration! (Using Misc Payload for a second battery)

Frame Weight: 1356.3g

Battery Weight (4S 5000mAh): 534.0g

Misc. Payload: 843.7g

All-Up Weight: 2734g

Flight Time in Hover: 10:00

Maximum Take-Off Weight Test Performed at 3190g

Tests with the larger motors are awaiting arrival of stock, expected around the end of February. You can order your kit with either motor set but the 880kV versions won't ship until February 29th. 850kV orders and orders without motors will ship next Friday, February 24th.

NEW TEST with the 880kV motors: (hovering in gusty winds)

Frame Weight: 1508g

Battery (4S 5000 mAh): 531.2g

Payload: 986.3g

Flight time: 10:17

All-Up Weight: 3025g

 -- only about 140g extra payload for the same duration flight but conditions were considerably more turbulent for the 880kV test flight

Frame Kit: $199.99

Frame + Motors: $419.99

Frame + Motors + Electronics: $649.99

Wow. Florian did his own motor controllers, RC (via Xbee) and everything else. From Hackaday: 

We think [FlorianH] did a bang-up job of prototyping his Minima Quadcopter on the cheap. The total bill comes in right around $200 and we’re very happy with the quality of parts as well as the results.

Here you can see the top of the double-sided board which he etched to host all of the components. At each corner there is a power MOSFET which drives the brushless motor. At first glance we thought that the Xbee module was acting as the radio control and processer as well. But on the underside you’ll find an ATmega32 which is responsible for reading the Gyroscope sensor and Accelerometer, processing these signals and driving each MOSFET via PWM lines to provide stability.

You can see some flight tests after the break. [FlorianH] mentions that there is some oscillation in the feedback loop when both the gyro and accelerometer are used. But cut the accelerometer out of the equation and the platform is rock-solid.

This build uses carbon tubes to mount the motors, which we think will be a little more robust than the all-PCB designs are.