Late last month drone company Matternet demoed a first, a precise and fully automated drone delivery (coffee!) to the roof of a Mercedes, capping a lengthy flight over the city of Zurich in Switzerland. Ardupilot is under the hood of the parachute equipped, 10kg, 70km/h copter.

From Techcrunch:

"Drone delivery may be closer than you think: Daimler’s Mercedes-Benz Vans teamed up with e-commerce startup siroop, and drone logistics provider Matternet to kick off a pilot in Zurich of its van-based drone delivery concept. The partners demonstrated a delivery live to an audience of media today, showing a coffee delivery from a small Swiss merchant using siroop’s platform."

More ...

In case you missed it, some great news items in the last couple weeks, from the Ardupilot development powerhouse:

- Release of Copter 3.5

- The Solo lives on

Copter 3.5 is a major release with an extensive list of new features and improvements, including support for visual odometry, obstacle avoidance, RTK GPS, dual GPS blending, non GPS loiter and navigation, improved autotune.

More specifically, AC 3.5 new features and improvements include:

• Performance improvements including 8kHz IMU sampling for improved vibration resistance, greatly reduced latency between IMU updates and motor outputs
• AutoTune with position hold if entered from Loiter/PosHold flight modes (wiki28)
• Delivery improvements including Servo Gripper17 and Package Place
• Driver additions and improvements:
  • Improved Smart Battery support (individual cells voltages, battery serial number, Maxell battery driver) (wiki25)
  • OLED Display51
  • Lightware LW20 lidar37, all lightware products can now use I2C interface5.
  • TeraRanger Tower29 lidar for Object Avoidance
  • Maxbotix sonar serial interface now supported
  • Garmin Lidar-Lite v337
  • Landing gear retracts/deploys on startup only after pilot moves transmitter switch
• Flight controllers supported now includes:
  • The Cube30 (aka Pixhawk2)
  • Beagle Bone Blue26
  • improved Intel Aero RTF support (ArduPilot is included in Intel's next software release)
  • AUAV2.1
  • PixhawkMini support
• GPS improvements:Mission fix for Turn-towards-next-waypoint-too-soon issue (only applies to AUTO flight mode)
  • Here+ RTK GPS33
  • Piksi Multi RTK GPS18
  • Dual GPS Blending (wiki30)
• Non-GPS loiter and navigation:
  • support StereoLabs ZED camera (visual odometry) (wiki22)
  • support Pozyx beacons (wiki11)
• Object avoidance:Precision Loiter using IR-Lock (wiki18)
  • normal (i.e. non 360deg) range finders supported (video27)
  • simple object avoidance in AltHold mode (wiki22)
• Pixracer10 improvements:Solo supported (we recommend upgrading flight controller to a Green Cube23 from a HEX reseller8)
  • ESC calibration fixes
  • board LEDs now work

More information on Arducopter 3.5 can be found here.

Copter  now also supports the Solo, a first. Detailed information can be found here: The Solo Lives on

Also stay tuned for the upcoming release of Arduplane 3.8, whose list of new features and improvements promises to be just as impressive as those of its Copter sidekick!

Great video from Ardupilot Development Team member Jaime Machuca, on working towards  an upcoming Maker Faire in San Francisco's Bay area. The Ardupilot event is sponsored by Intel.

More  information on the project is available here.

Every year Google proposes to fund students for the summer with its GSoC program, where open source organizations and students are announced  through a highly selective process. This year was the first time Ardupilot applied as its own entity, and Google just granted it  four student  positions. No doubt due in part to the high number of student proposals submitted to Ardupilot, and the high quality of many proposals.

This is great news for the drone and robotics community, professional companies, DIYers and professionals alike, and the open source world.  With lots of new development in perspective, adding to the already fast paced and steady stream of Ardupilot development and innovation. 

Congratulations to the student accepted, who will be individually mentored through the summer by some members of the development team. More information about the proposals and selected students can be found here.

In which a 1.7m flying wing VTOL transitions to forward flight with ... oops ... reversed elevons. As the transition starts the quad motors are active and compensate for the reversed elevons. Once the transition is completed the quad motors stop and ... the flying wing, now flying like a plane, flips 180 degrees quasi instantly due to the reversed elevons. The VTOL heads straight towards the ground.

But ...   Read more ...

Ardupilot developer and Mission Planner author  Michael Oborne  gave a talk at the Ardupilot 2017 Unconference about little known features of Mission Planner. A video of the talk is available, with Michael discussing  items including APSync, FFT (Fast Fourier Transform) for vibration analysis, a fast method to draw polygons for mission planning, moving groups of waypoints on the map, and swarming.

There is also a video featuring a  flight demonstration of swarming, a feature recently added to Mission Planner.

Interested readers will also catch a glimpse of  StopRotor's  Rotorwing in the video, a helicopter and plane hybrid prototype displayed in the conference room.

Read more ...

The Ardupilot Development Team held an "un-conference" (aimed at developers) in Canberra, Australia last month. The conference included several talks, along with flying in the field.

Ardupilot developers present included  well-known Plane, Copter and Rover  leads Andrew Tridgell, Randy Mackay and Grant Morphett, and developers Paul Risenborough (EKF), Leonard Hall (control and auto-tune),  Philip Rowse (Pixhawk2),  Michael Oborne (Mission Planner).

For those interested, the first two talks and demos  videos are available on Ardupilot.org following these two links: 

Talk and demo by Andrew Tridgell, Talk and demo by Paul Risenborough.

Great news for the open source  community and the greater world of autopilots: On February 27th Ardupilot.org was selected as a participating 2017 Google Summer of Code organization.

Sponsored by Google, GSOC is a unique program matching open source organizations and students interested in participating in open source development.

Students: Spend your summer break writing code and learning about open source development while earning money! Those accepted  in the program will work with a mentor,  working directly with world class programmers. Many past GSOC students also became lifetime open source developers!

Google Summer of Code is open to post-secondary students, age 18 and older in most countries.

Those interested can find out  more information about this unique opportunity here.

From Rusty and Jacob of ArduSub:

Hello everyone, Rusty here at Blue Robotics. We're pleased to announce that the ArduSub project has merged with Ardupilot!This is a momentous occasion for the ArduSub project, with our two main developers, Jacob and Rusty, both becoming members of the ArduPilot development team. ArduSub is the first new vehicle type since the addition of ArduBoat in 2011, and is the first to take the ArduPilot project underwater! We've been looking forward to seeing this since the start of ArduSub!

Read more here ... 

From Tridge at Ardupilot.org:

We just finished the ArduPilot Developer meetup in Canberra, Australia. Randy [Lead Arducopter developer] will be posting videos from the talks when he gets back to Japan, but meanwhile here is a nice video by Darrell Burkey of some of the activities over the weekend. ... Read more

Among other glimpses of goodness the video features the successful and  rather spectacular, IMO, example of  the Addiction X plane ( prior simulation work referred to here) hovering vertically on its  motor and  entirely controlled by  autopilot.  ( "Hands off" controller sticks, no pilot manual intervention). 

The AddictionX is a popular model plane allowing for spectacular high precision acrobatics. In addition to more common barrel rolls and vertical loops, its extremely light  wing loading and high motor thrust to weight ratio provides expert flyers with  capabililies such as precise vertical hover and  knife edge maneuvers.

What if this kind of flying could be performed by an autopilot? Lead Arduplane developer Andrew Tridgell and the Ardupilot development team got started towards doing just that. Building up on previous  VTOL work such as  Ardupilot support for quadplane, tail sitter, tilt rotor and tilt wing VTOLs along with  fully autonomous inverted flights, Tridge  interfaced Ardupilot to the popular RealFlight flight simulator, and successfully flew the simulated AddictionX model, including vertical hovers, with Ardupilot.

Coming up in the near future: Tridge actually owns an AddictionX, and expects to fly one with Ardupilot. Also coming up with Ardupilot:  ArduSoar, a  soaring "extension" of ArduPlane for gliders. Stay tuned!

Read more here including video  here  including video.

An eye opening and entertaining post over at ardupilot.org from developer Grant Morphett:

"Some of you may be wondering how much real world testing us developers really do. Well the answer is we try to test as much as we can. We especially try to test the limits our boundary conditions that most people won't test because they like their vehicles in one piece.

This thought process led me down the track of wondering if I could do a fully autonomous landing of my test plane - UPSIDE DOWN. i.e. flick the plane into inverted flight via the switch (howto - http://ardupilot.org/plane/docs/inverted-flight.html4) and let the landing software run its course. A quick test in the simulator and it worked great. So on the weekend we made the above modification to my test plane which is currently a FirStar EX.

Read more ....

And a great picture of the winner of the Airbus/Local Motors Cargo Drone challenge, a quadplane, prior to its successful maiden flight running Arduplane. (More information about it here  earlier on this forum, and  here)

http://www.open-electronics.org/the-most-advanced-autopilot-software-ardupilot-is-the-answer/

Great feature coming up soon in Ardupilot's Copter 3.5. And congrats to the contribution of the month winner!

...

Congratulations to Peter Barker for winning the "Best Contribution of the Month" for Dec 2016 for his Package Place Feature (video here).

This new mission command (which will be released with AC3.5) allows a vehicle to deliver a package autonomously without the vehicle actually touching down. It accomplishes this by carefully monitoring the motor level until it detects the package touching down. It then opens the gripper to release the cargo. This is very handy especially for delivering packages on surfaces (i.e. mountain sides or the sea!) where the vehicle cannot safely land. Read more, including videos,  here ...

Many were skeptical after the first video, including me.  Yet  my skepticism lost some ground when I looked further into the capabilities of Jetcatusa micro turbines.  A few weeks ago after that first flight, he announced an upcoming major public demo later in April, when interviewed by a French magazine ... 

Well, he delivered.

2.2 km. 

This is huge, imho.

Earlier video and related thread here.

To my knowledge the first time a "major" news publication picks this up ...

From the 3/11/16 Edition of Newsweek Magazine:

The U.S. House Energy and Commerce Committee heard testimony in November on whether and how to regulate unmanned vehicles. This drone did not testify.

Full article here

Affected drones can't be patched, researcher says

Also a  bit difficult to  understand how that  "police drone"  could have been sold for $30,000-$35,000. Looks like run of the mill,  Tarot type frame, motors, etc ... X8, $2,000 or so at most?

Eagerly anticipating the great work being done on Mavlink2 at DroneCode!

Great application leveraging DJI's SDK and IOS/Android vision processing. Just signed up as a beta tester.

After Randy's pioneering work with his balloon finder on Pixhawk, can't wait for something similarly integrated with Dronecode, OpenCV,  ROS ...

A very interesting post over at the Phantom Pilot forum.

From ianwood:

I had my first ever bonafide flyaway about 6 weeks ago. I had my theories, bought $1,000 in replacement parts, rebuilt and moved on. Then it happened again.

This time, however, I had a data logger attached to the Phantom, a custom made miniature "black box" that records 75+ different flight parameters 5 times per second. I've spent the past day analyzing the data and I've discovered some things that I didn't suspect before. 
  
  
  
None of the sensors including IMU, compass, GPS, etc. were to blame. The battery, ESCs, motors did exactly as they were supposed to do. There was no loss of connection, GPS interference or any other external factors.

The NAZA locked up. It had intermittent freezes that stopped it from functioning normally. Here is a snapshot of the logged data from when the NAZA stopped working normally.



At 12 minutes, 45 seconds, the data shows 5 anomalies:

• The Course Over Ground (COG) freezes. It had been fluid all throughout the earlier part of the flight.
• The speed (Total SPD) changes to 769,361.92 m/s. It had been normal throughout the earlier part of the flight.
• The satellite count (SAT) jumps to 204 then returns to normal.
• Erratic updates of ATTI pitch and roll values. 2Hz or slower.
• Erratic updates of motor speed. 2Hz or slower.

The cause is likely to be one of two things:

• Defective hardware in NAZA (e.g. cold solder joint) that took several flights of light vibration to manifest.
• Firmware defect when flying in edge condition (corner of the envelop) that rarely manifests.

I am working on recovering the video (file is corrupted due to loss of power) and I also have a 3D "fly through" of the incident which I will post later.

I've included a more detailed analysis of the data which is in the PDF link here: http://www.ianwood.com/docs/anatomy-of-a-dji-flyaway-v1.pdf

The recent 2015 Embedded Linux Conference held March 23-15 in San Jose, CA featured some great presentations related to drones. Videos of those talks are now available, and it's been a great pleasure to watch them.

Here is  an overview for easy reference, with a short description (my personal take) for those of us interested who were not able to attend. Be sure also to check the link above, many presentations are not covered here, yet could be of interest.

Cheers to all speakers, the DroneCode and Linux foundations, 3DRobotics, and the companies sponsoring the foundations and speakers!

-Olivier Brousse

----------------------------------

Keynote - DroneCode Project and Autopilot With Linux, by Andrew Tridgell: Slides, Video

In this talk, master open source developer and uber hacker (Samba and  Rsync, prior to Ardupilot)  Andrew “Tridge” Tridgell  presents  the history of the open source Ardupilot flight controller, from the beginnings of Ardupilot  to current applications and the foundation of DroneCode.  He covers the Arduino world and the open source community, the creation of this  forum by  Chris Anderson, through to his involvement in Ardupilot and the subsequent  hardware and software evolution. Tridge then presents the many applications in which Ardupilot is being used, and discusses the structure of Ardupilot code and its ports to Linux.

Autonomous Navigation for an OMAP4 Nano-Drone, by Grégoire Gentil: Slides, Video.

In this presentation Gregoire Gentil of Always Innovating presents the nano drone “Dronie” (aka MeCam), a quadcopter capable of autonomously following a person using computer vision. The Dronie’s flight controller is based on Texas Instruments’ OMAP44x board, boasting  a dual Cortex A9 kernel, a dual cortex M3 running TI’s Real time operating System, and a DSP used for computer vision.  Along with a 9 DOF sensor and a sonar, up to five cameras can be aboard the tiny drone. After a general hardware presentation, including the basics of how drones fly, Gregoire demos the drone following himself (no GPS!), and discusses the computer vision software algorithms that he developed. He concludes the talk by discussing sensor fusion details and the challenges he had to overcome.

If you are relatively new to drones, curious about “what’s under the hood”, and only have time to watch one or two of these presentations, this technical and entertaining talk, along with Tridge’s keynote, may be the one for you.

“Open Source Drones on Linux”:  Lorenz Meier, Slides, Video

In this presentation by drone pioneer and ETH researcher Lorenz Meier , the founder of the hardware and software open source PX4/Pixhawk  flight controller platform  presents a retrospective of the work he was involved with, and what’s ahead with the DroneCode foundation.  Lorenz, who is also behind the Mavlink  protocol and the  Qgroundcontrol  ground station software, starts out presenting the long history of Linux used in robotics and drones. He then discusses details of the PX4 platform and its software architecture, details current challenges and limitations, and provides information on the future of PX4. A great Q&A session follows, including further discussions on the second release of the Robot Operating System  (ROS2) , the UAVCan protocol, wireless link security considerations, and DDS to augment the capabilities of Mavlink.

“Drone API: A Tutorial on Drone Control”: Kevin Hester:  Slides, video.

In this talk, software and Ardupilot developer  Kevin Hester presents  DroneAPI, the  application development toolkit  allowing programming of simple 'high level' drone operations (e.g. move the drone to this position, perform that task there, etc ...). He shows how easy it is to get started, setting up “live” the development kit on a Vagrant virtual machine environment. He then runs some programming examples featuring Ardupilot computer simulation (Software in the loop, or SITL) , discusses  how they can be used further, and introduces the cloud-based app  DroneKit.

 “Application of Data Fusion to Aerial Robotics”, by Paul Riseborough: Slides, Video.

Here the Extended Kalman Filter master and Ardupilot developer Paul Riseborough covers the EKF which he developed for Ardupilot based flight controllers. Paul discusses the limitations and challenges of off-the shelf gyroscopes and accelerometers needed to control drone stability, covers sensor fusion, and provides a clear high level overview of why the filter is needed and how it operates.

 “Flying Penguins: Embedded Linux Applications for Autonomous UAVs”, by Clay McLure: Slides, Video

In this talk Clay McLure recounts his recent experience in the world of drones, where in just six months he built and flew his first drone and successfully developed onboard code for precision landing.  In this wide ranging presentation Clay  discusses the basics of ROS (Robot Operating System) and how he used it with Pixhawk via Mavlink and MAVROS.  He then details the code that he developed and how it runs on a companion computer.  

Using Intel Edison to Fuse Embedded Linux With Existing Drone Flight Controllers, by Mark F. Brown:  Slides, Video

In this presentation Intel engineers Mark Brown and Joel Rosenzweig present their implementation of a hardware and software environment allowing for autonomous drone flight,  on top of an existing manual (non-autonomous) flight controller treated as a “black box”. Starting with a DJI Naza running proprietary code, they use an Intel Edison running Linux and build a Node.JS web server to allow for setting waypoints over Google maps. The Edison is also connected to an additional GPS and IMU to control the Naza flight controller. This is implemented with the help of three Atmel A328 micro-controllers providing the “glue” between the Edison and the sensors. A Parralax P8X32A board then multiplexes serial commands from the Edison and RC inputs, and outputs PWM to the Naza. An impressive endeavor which was completed after hours in under eight weeks!

The Syria Airlift Project: Open-Sourcing Humanitarian Airlift, by Mark Jacobsen: Slides, Video

In this presentation US Air Force pilot Mark Jacobsen, founder of the crowd funded  Syria Airlift Project, presents his project, which “seeks to use humanitarian drones to deliver life-saving aid to besieged and hard-to-access populations in Syria”.  No doubt familiar to many readers of this forum, this great endeavor is a beautiful example illustrating how drones can be used for the greater good. If you find this project worthwhile please consider contributing to the Indigogo campaign!

Teaching More Fish to Fly,  by John Hawley:  Video

In this talk John Hawley , Open Hardware Evangelist at Intel, first discusses the rise of small drones from its DIY/Makers origins to the commercial world.  He presents his use of the powerful  Minnowboard Max as an auxiliary computer that can be used for vision processing, allowing features like collision avoidance. The Minnowboard is connected to the Pixhawk flight controller taking advantage of Pixhawk’s serial interface, similar to the way a Rasberry Pi or Odroid have been connected and used. John then discusses applications where auxiliary computers are used. 

Heterogeneous multi-core architecture support for DroneCode, by Mark Charlebois: Slides, Video

In this highly technical Linux talk Mark Charlebois of Qualcomm discusses porting the PX4  flight controller software to the Qualcom’s SnapDragon 600 hardware platform.  Mark first introduces the SnapDragon board with its four Krait cores and Hexagon DSP and discusses the board’s overall hardware architecture. He then presents his approach and work porting the PX4 software environment from its native Nuttx real time operating system to Linux, running on the SnapDragon. If you are a Linux programmer new to the PX4 flight stack this talk is for you!

Additional Talks

While not specific to drones, I also found the following enjoyable and informative:

 “Embedded Linux moves into High School Robotics”, by Michael E. Anderson: Slides, Video

I enjoyed watching this talk introducing the world of robotics in high school environments. Presenting the FIRST (For Inspiration and Recognition of Science and Technology) Robotics competitions, Michael Anderson introduces and details the RoboRIO Robotics controller, shows some example code, and shares his experience overseeing student teams.

10 years of open-source robotics, by Laurent Pinchart: Slides, Video

A great retrospective by Laurent Pinchart of his 10 years of involvement with the Eurobot Robotics competitions occurring in Europe every year. 

Creating Open Hardware Tools,  by David Anders: Slides, Video.

In this presentation David Anders of Intel reviews the history of open hardware tools used to debug hardware environments, from the humble multimeter in the beginning to logic analyzers and specialized boards necessary to debug today’s complex hardware platforms.