Blogs

My video editing skills leave alot to be desired...

 Decided to write up about my experience in building my lastest multicopter, a 3dr hexa. The lessons learned from the exercise include

1. Check before you start , that you have all the screws  and spacers are available.
2. Decide on what flight controller you are planning to use.
3. Start your build early enough, so that you could catch some sleep.

 I should thank the 3dr folks again for the awesome work they have done. The build was straight forward, the only challenge was that once i opened the box, and laid out the contents, I realised that some nuts and spacers were missing. This lead to my next lesson, given i had promised myself to take the hexa for test flight next day, my choices were limited, cannibalise an already assembled 3dr quadcopter and hexacopter ( from quadframe) or not to build. So I decided to proceed and strip my 3dr (from fully build to mere frame and bits missing) and took my (qaudframe) hexa to parts.

The next one to make my mind up on which flightcontroller to choose. I had a choice of six in front of me, mk,uavp, st32, APM1.4, hoverfly and APM2.Again, this goes back to lesson number 3, the later in the night the hazier your brain becomes. Finally i decided to go for the APM2. The final result of this build defenitely shows the signs of non functioning brain.

Yes, you can major in "drones" now in college. 

From Bloomberg (via sUASNews). Excerpt:

Russell Peters is a junior at Embry-Riddle Aeronautical University (78211MF), one of the top U.S. flight-training schools, who has no intention of taking to the skies.

Standing next to a 3-foot-wide (91.4 centimeters) plane that searches for objects on the ground using artificial intelligence, he talks about how his fascination with robotics led him to theDaytona Beach, Florida, campus to prepare for a career building and operating drones.

I don’t care about flying the planes,” said Peters, of ColoradoSprings, Colorado. “I just want to make the planes. And if we can get rid of the pilots, that’s great.”

As the U.S. Federal Aviation Administration prepares to let civilian unmanned aircraft operate in domestic airspace, universities including Embry-Riddle have created majors in flying and building drones. Enrollment is accelerating as students look for new opportunities in an aviation job market pummeled by airline bankruptcies.

The drone industry, estimated worldwide at $5.9 billion annually, will expand to $11.3 billion by 2021, according to areport last year by the Teal Group Corp. of Fairfax, Virginia, which analyzes the industry. It’s “been the most dynamic growth sector of the aerospace industry this decade,” the firm said in the report.

...

Alex Mirot, a retired Air Force major who piloted Predator and Reaper unmanned vehicles built by General Atomics Inc. ofSan Diego, coordinates Embry-Riddle’s degree program inUnmanned Aerial Systems Science. In less than one year, 75 students have enrolled to major or minor in the subject, he said.

'Kind of Amazing'

“To me this is kind of amazing,” Mirot said in an office festooned with photos, ribbons and other memorabilia from his 10 years in the military. “That’s how popular this is as a technology.”

The program offers two tracks for students: training on simulators to be drone operators, or non-pilot roles such as operating remote-controlled cameras and managing unmanned- aircraft programs.

Students can take classes in how unmanned systems work, robotics, business management, operating remote sensors such as cameras and radars, and drone simulator training, Mirot said. They must take courses in electrical engineering and programming, he said.

Across campus at the Department of Mechanical, Civil and Engineering Sciences, students like Peters study how to build and design unmanned aircraft, Charles Reinholtz, the department chairman, said in an interview.

Graduates Getting Jobs

The University of North Dakota in Grand Forks established a drone-study major more than two years ago and has graduated 17 students, Benjamin M. Trapnell, an aeronautics professor who developed the program, said in a phone interview.

Graduates can expect to make starting salaries of at least $50,000, and perhaps double that if they accept jobs in dangerous locations, Trapnell said. That dwarfs entry-level salaries as airline pilots, which can be below $20,000, according to government data.

He said his graduates, who must obtain FAA commercial pilots’ licenses, have landed jobs with drone operators and manufacturers such as the Aeronautical & Space Systems division of Information Systems Laboratories Inc. of San Diego and AAI Unmanned Aircraft Systems, which is owned by Textron Inc. (TXT) of Providence, Rhode Island.

It’s premature to design courses and pilot training programs for unmanned aircraft because the government hasn’t yet outlined requirements for how they must be built and flown, said the head of the only U.S. civilian test bed for unmanned aircraft.

No U.S. Standards

“We’re having a hard time understanding how you develop a bachelor’s degree program when you have no federal standards,” Doug Davis, who directs the drone program at the New Mexico State University (28344MF) Physical Science Laboratory in Las Cruces, said in a phone interview.

Gene Gamble, an Embry-Riddle senior, is on a team led by Peters that plans to enter a national unmanned-aircraft contest later this spring. He said he relishes seeing the amazement of people who stop by the drone lab for tours.

“I do enjoy telling people that I work on robots that drive themselves,” said Gamble, who’s on a five-year program leading to a master’s degree in software engineering.

Gamble spoke while replacing the battery on a winged device slightly longer than one foot that hovers like a helicopter with one blade. The craft is based on the way a maple seed rotates as it falls from a tree.

Sensor Design

Using a hobbyist’s radio remote-control handset, he flew the device more than 10 feet up into an atrium. Gamble is helping design software and sensors that will eventually allow the device to fly and navigate itself.

“I’d like to work for a company that does things like this,” he said.

Michael Tagliereni, of Jackson, New Jersey, an Embry-Riddle senior majoring in aeronautics, said he enrolled in an introduction to unmanned aircraft class as a fallback in case he can’t find a job as a pilot or an air traffic controller.

Sitting next to him in a classroom, Paul Renart, a freshman from Warwick, New York, said he chose to major in drone studies because it will help his military career. Renart is in the Air Force Reserve Officer Training Corps.

For Peters, the junior drone-building enthusiast, his fascination with making intelligent machines has put him at the forefront of a field with a large measure of cool on campus.

“I like to be on the cutting edge,” Peters said. “I think it’s just the beginning.”

Update from Hackaday. This looks like a good candidate design for a cheap open source version of the cool U-Penn GRASP Lab swarming quads.  The company plans to manufacture and sell them. 

Crazyflie, the itty bitty quadcopter that uses a PCB for its frame is at the cusp of being able to fly without a PC. We were enamored with it when we first spotted it, and the Crazyflie has only gotten better. As you can see, a new PCB gives it a slightly more stylized look, but that’s not all. There have been major improvements to the design of the software, and various parts of the hardware. They’re waiting for the final prototype to arrive from the factory to test the ultimate goal, PC-less flight.  We’re not sure why this tiny quadcopter is so enthralling, but it sure has us captivated.

Lost control of my Quad a couple months ago and it landed in a nearby pond, me and my kid(pictured) spent like 20 minutes in a cold 4 ft deep body of water looking for it. I really thought that I had lost everything. We finally found it and it was still beeping, we pulled off the battery plug immediately and just could not believe how in the world this happened. Anyways, got home and placed all the electronics in a container full of rice to suck the humidity off, after a few days we decided to turn the power on and see what kind of damage had occurred, to our surprise the only thing we had lost was 2 ESCs..(very lucky)..  

Hey everyone, it always bothered me when my Quad hit the ground hard, so I started looking on a way to add a suspension to it and after 2 weeks of trying, this is what I came up with:

I got some shock absorbing from a 1/10 scale buggy and added wheels, one on each side of each leg with a total of 8, they are very light (used on some airplanes like Apprentice etc). Had to 3D print some parts to hold the shocks on the Arm, I also printed the Sonar support. I am very happy with the result of the suspension. To test it I cut the power off at about 2 feet high and it hit the ground but it doesn't bounce at all (the shocks absorb all the impact). I'm very satisfied with the results, and I just wanted to share.

On 4/14/12 we flew an APM2 ardupilot in a balloon to over 100000 feet for 7 1/2 hour 300 mile + flight !

The flight originated near Albuquerque NM and landed near Trinchera ,CO. The flight was over 300 miles and lasted far longer than we planned. The flight was so long the 5 on board cameras ran out of memory and battery before the flight ended.

The chase crew went 225 miles before giving up because there was no data from the Digi M10 Satellite modem. The crew was almost back into Albuquerque when all of the sudden three phones got e-mails from the Digie modem at the same time.

ET CALLED HOME !!!

It gave us the Lat, long, and altitude of where it was. Several emails later we realized the package landed. We quickly put the coordinates into google maps and could not believe where it was ! It was in Colorado! And it landed 7 1/2 hours after launch !

I will download the data from the APM2 SD card and see the flight.

Here are some pictures.

Could have landed here in Colorado

But it landed 10 miles further 50 feet from the edge of a farm road !

Some of the equipment flown...

I will post more pics and some videos later when I get back to Albuquerque. My car motor seized up out in the boonies right after retrieving the package. Got to get rental car and find way to get new motor in 2006 Saturn Vue.

Earl

This is spring time and thus the ThermoPilot project is still under intensive development (now the v6.x version) and real tests flights can be conducted because the thermal upwards can be hunted above the fields... Below a sample log flight during a full autonomous thermal hunting in free mode.

 Here a flight log during a 23 km of full autonomous cross country soaring mission in closed loops with 5 waypoints.

An ArduPilot Mega 2560 is installed on the Cularis e-Glider with the ThermoPilot v6.x firmware (an enhanced version of  the ArduPlane v2.33). The ThermoPilot v6.0 includes a lot of improvements for the cross country soaring and the thermal hunting and some dedicated ML algorithms to do a close following of the thermal upwards.

Below the Cularis e-Glider and its big brother, the Fashionista: a 5 meters wingspan e-Glider (a very good thermal hunter)...

During the summer and the spring 2012, the ThermoPilot project will be very active in the air... Stay tuned...

Regards, Jean-Louis

There are a lot of home-brew parts in this machine (no fooling, really).

I had orignially fabricated parts based on the 'A' frame but after noticing the 'B' release, I ordered replacement parts and modified mine as close as possible. The bottom plate is 1/8 aircraft plywood painted. The top was made from an FRG piece I had in a junk box.

Getting access to the PDB is now much easier and if you have an 'A' and break it, go to the 'B'.

The legs are made on my home-brew CNC Dremel router.

Using the posted 3DR drawings, I made some G-code that cloned the 3DR leg pretty close. I didn't bother with the nice radius at the 'foot' end but may tinker with the code later. The harvested circuit board material found new life rather than being tossed in the landfill to decay over a thousand years. This way I get to break it in spectacular fashion!

I used 6-32 screws and remachined  some aluminum hex standoffs I had for the leg hardware. I let the CNC router crank out two complete sets of legs as I suspect a few will get broken while I earn my quad wings.

Using what I have in my shop/garage and buying significant parts from 3DR has been my approach.

Things left to do/make are:

• Electronic stack plates - I may go with 1/8 aircraft ply. I have to make G-code for this.
• Optimizing the wiring and completing the connectors
• Weight reduction - perhaps some material off the arms beyond the leg mounts?
• Ebay stuff to fund the APM and the rest of the electronics.

More to come!

April 14, 2012

I've been assembling the PBD, ESC and motor connectors.

I went with the XT60 for the battery side rather than the Deans.

After routing the motor wiring as mentioned previously, I added another layer of heat shrink to the wiring to provide another level of anti-chaffing.

For those of you interested in the actual hole pattern on the arm...

As this is my trainer quad, as mentioned before, I think the next arm will have radiused corners out past the motor. This will reduce impact edges and give the arm a more finished look I think.

Here is one of the arms with the extra heat shrink at the motor, the ESC with all connectors zip-tied in place.

I took the Arducopter, prior to mounting the just shown wiring updates, to one of the local shops in Greenville, SC,  The Great Escape. Keith is the resident heli-nut and we discussed the need for me to spend time with a simulator. Real Flight 6 Heli now has a quadrotor in the selection. The in-store demo unit had me hovering and breaking props in between reset button presses. It looks like a great product and at rev 6, it seems that it won't be going away anytime soon.

Keith inquired about the PDB and had not heard of 3DR. A quick trip on one of the store's PCs to the DIYDrones Store and he was in love with the Hex copter. I tried to explain the APM and its open source development. He was accustomed to a single threaded application (dumb controllers)  rather than a 'smart' controller like APM.

We had a good visit for about an hour or so and I promised to return when the Arducopter was getting closer to flying.

Before I get the funds for the APM order, I may build a servo 'stimulator' and  check the motor directions. Anyone ever try that before bolting in an APM?

May 19, 2012

Checklist items acquired for the quad:

• Spektrum DX6i (with DSMX) an ebay win
• AR6210 receiver with sat receiver for above
• Cable allowing the DX6i to operate FMS sim software
  • Been practicing with an FMS quad model - the only problem, it does 3D! INVERTED HOVERING!

Still building funds for the APM2 purchase. I gathered parts for the servo stimulator and need to get that sub project going. I may just breadboard it and then used one of my favorite programming languages - solder.

Quick video of my newest bird, an Xaircraft X-650 V-8 featuring all new flight controller and AHRS V.2.  I didn't realize I was in Acro mode in the beginning as seen in the video but figured it out eventually.  Must have been some English/Chinese translation issue or maybe just the fact that I didn't read the manual :)

Anyway, enjoy the video

Hello guys,

this is my first post on this blog and I'd like to give some information about my new personal project. How you read on the title, the project is a RC CAR with Arduino nano that is controlled by TB6612FNG.

The final project will be a RC CAR controlled by Arduino nano and a few sensors, the big goal is to create a car that I can setup a route and it should be cleaver enough to avoid the obstacles from point A to point B.

I don't like to use the Ardupilot because I want to be proud of myself and create all the AI by myself :)

Tonight I just soldered the driver shield to the main motor and then I tested it.

Items:

http://www.sparkfun.com/products/11113

http://www.sparkfun.com/products/9457

http://www.amazon.co.uk/gp/product/B002GOHOA2/ref=oh_o00_s00_i00_details

The next items that I should to buy:

http://www.sparkfun.com/products/8130 or http://www.sparkfun.com/products/10616

http://www.sparkfun.com/products/10472

2x http://www.trossenrobotics.com/parallax-ping-ultrasonic-range-sensor.aspx

2x servos

This is my personal blog where I am going to publish the other projects with Arduino: http://projectsmax246.blogspot.co.uk/

One of our priorities as 3D Robotics grows is to improve the customer experience in everything from product design and documentation to packaging. You'll see a bunch of new efforts in this direction over the next few months, ranging from a branding/design overhaul of the storefront to a new platform for our documentation that should be easier to use and better looking than the current wiki. We're also going to start shipping APM in a professionally designed enclosure, which we'll preview next week. 

Here's just one small example: our new boxes!  I think this is the first public use of our new logo, too. We particularly like the Easter egg of the rudder replacing the dot of the i ;-)

One of the APM features the dev teams have been working hardest on is improved compass calibration. Magnetometers are a tricky thing to get right because there are at least three kinds of variation: 1) each sensor is different, 2) the Earth's magnetic field varies around the planet (Declination), 3) The "hard iron" magnetic interference of each aircraft is different, and even that varies in flight when the motors are going. 

Right now we have three different kinds of calibration you can use: 

1. Auto calibration. You do nothing, and the code figures out all the offsets and declination by comparing the compass readings with the GPS and IMU readings over time in flight. Pro: no user effort. Con: it takes a few minutes of flying to get right, so the compass is inaccurate at first launch.
2. Manual calibration in the Mission Planner. This is the screen on the Configuration screen, where you can enter your Declination and then press the "Calibration" button and move and rotate your aircraft around for 30 seconds while it records the data and does some math to calibrate the sensor. Pro: works. Con: it's a little awkward, especially for big aircraft. Also doesn't reflect the magnetic interference that can occur when the motors are going in flight.
3. Replay a flight log. This is a very cool option, shown above, where you can just replay a flight log (.tlog) and the code will compare the GPS and IMU readings with the compass reading and make the necessary corrections. Pro: works great. Con: must have already flown, will be messed up if you load a .tlog file where you didn't actually fly. 

It's interesting to see how others do this. In the video below, from the radio program Marketplace, you can see (at 1:20) how Foxconn calibrates the magnetometer in an iPad. It's basically the same as our Method 2, but highly automated:

Something a little different from the TBS Headquarters :) Hope you guys enjoy!

I can't believe just how wrong  John Villasenor can be, I guess he figured one of us would mention this here and it would drive his blog posts wild at Scientific American.

He completely misses the real definitions of UAS its not about having an autopilot onboard. What worries me more is that Mr Villasenor is asked to speak about UAS to other grown ups. There is a huge gap between the reality of UAS and what CSI tells us.

John is a member here so he can perhaps chip in and explain his definitions.

From the FAA Interim Guidance http://www.faa.gov/about/office_org/headquarters_offices/ato/service_units/systemops/aaim/organizations/uas/coa/faq/media/uas_guidance08-01.pdf

Unmanned Aircraft: A device used or intended to be used for flight in the air that has no onboard pilot. This includes all classes of airplanes, helicopters, airships, and translational lift aircraft that have no onboard pilot. Unmanned aircraft are understood to include only those aircraft controllable in three axes and therefore, exclude traditional balloons

Here is an excerpt from the Scientific American blog.

What is a "first person view" unmanned aircraft? Is it a drone?

http://blogs.scientificamerican.com/guest-blog/2012/04/12/what-is-a-drone-anyway/#comment-4952

A first-person view (FPV) aircraft has a front-facing video camera and transmits real-time video to an operator on the ground. The operator looks at the image on a computer screen, sees the view as if he or she were sitting in the cockpit, and flies the plane accordingly.

While “UAV” is a general term for (non model aircraft) unmanned aircraft, FPV refers to the subset of such aircraft that are flown by a remote pilot using the image transmitted from an on-board camera. Unmanned aircraft guided exclusively by GPS or on-board computer analysis of imagery are UAVs, but they aren’t first-person view UAVs.

Until now, most FPV aircraft have been operated by the military, using technologies that make it possible to fly the aircraft beyond the line of sight of the pilot. However, use of FPV aircraft in non-military settings is certain to increase with the recent enactment of a new U.S. aviation law that will open U.S. airspace to many types of unmanned aircraft in the coming years. FPV aircraft are likely to be subject to very conservative FAA rules regarding domestic non-line-of-sight operation to minimize potential safety concerns. For example, if the communication link between the pilot and the aircraft fails, then there are obvious challenges involved in bringing the aircraft back to the ground without endangering other aircraft or people on the ground.

Is an FPV aircraft a drone? Under the strictest definition of drone, it isn’t, since it is flown under the control of a human operator. However, when flown beyond the line of sight, an FPV aircraft would be characterized by many people as a drone, despite the significant skill that might be involved in flying it. This is because the definition of drone can be difficult or impossible for an observer to apply. After all, how can someone who sees an unmanned aircraft maneuvering without any evidence of a nearby pilot know whether it is autonomous or remotely piloted? From the observer’s standpoint, it’s not unreasonable to consider it a drone.

Nothing much to do...

Do you know about the Internet Of Things?

Dragino MS12 is a wifi/linux enabled appliance for MCU projects. It uses the awesome OpenWrt.

The goal of the Dragino is to solve the connectivity problem and greatly enhance micro-controller products. So, it has an internal bus for a dedicated micro-controller board which defines the platform that would interact with the physical world.

So, Seeedstudio added an Arduino based daughter board inside it and came with "Dragrove".

"Dragrove is an open source based generic gateway for internet of things. It combines Dragino (Easy Internet Access, powerful CPU, Linux system), Arduino compatible interface board (monitoring and controlling the physical world), and RF networks like X-Bee. "

The term "Grove" on the illustration above is related to a product family with dozens of sensors and other modules for easy prototyping. BTW... I've proudly designed the logo. =)

At first look it sounds a nice product.
I don't need to think too much to find some cool applications for it. What do you think about?

Here is my newly built DJI Flamewheel F450 frame bought in goodluckbuy.com...

I was thinking to attach my APM2 on this quad or just let it fly with the FreeFlight Controller that came with.
I am still looking for battery packs to power this quad any suggestions will be much appreciated.
and still waiting for the availability of the Turnigy 9X in hobbyking or any variants of it that can be flashed/modded

all comments are welcome may it be from quad vets or new in the hobby
will be fitting this out with gopro and fpv or if it can carry a casio exelim z130 or my nokia N900

Unboxing:

Assembled and Wire management:

APM2 on DJI Flamewheel frame

DIY Drones Take on Silicon Valley

http://online.wsj.com/article/SB10001424052702303299604577326301981308414.html

This is the story I shot video footage for. Andreas Oesterer and Mark Harrison are featured.

And here is the full flight video:

I'm a big fan of Lego Mindstorms as an entry point for robotics and physical computing and even got my start in UAVs with it. 

Here's the most UAV-friendly sensor yet for Mindstorms ($85): 

The CruizCore® XG1300L is a digital fully self-contained MEMS gyroscope and accelerometer. It is compatible with the Lego® Mindstorms NXT kit. 
It includes a signal processing unit that estimates and compensates for the most significant errors that affect gyroscopes, providing an output with minimum error. For convenience, the CruizCore® XG1300L internally integrates the gyroscope output in order to compute the angle. The angle value can be used directly as the heading measurement in navigation applications.
The CruizCore®XG1300L includes a three-axis accelerometer that provides calibrated accelerometer output. The accelerometer can be used to compute tilt angles, detect robot collisions, or to determine changes in velocity.
The CruizCore® XG1300L communicates with the Lego® Mindstorms NXT through the I2C bus. It implements a communication protocol compatible with Lego® NXT, and users can use their favorite programing tools to communicate with this device. 

The sensor outputs from the CruizCore® XG1300L are near-free of errors and can be used immediately in applications.