Blogs

CHRIS ANDERSON – AMA MEMBER

Chris Anderson just left his post as Editor in Chief of Wiredmagazine to focus on his cutting edge UAV manufacturing company, 3D Robotics. Chris will be speaking about his role in leading the development of low cost UAVs using the extraordinary advances in smartphone based sensor, imaging, wireless communication and open source processing technology. He will share where the movement is going for small, non-commercial UAVs currently being operated under FAA’s recreational unmanned aircraft guidelines found in AC 91-57. Chris has authored three books including his just released book, Makers: The New Industrial Revolution.

January 11-13, 2013

Ontario, CA   <-- "CA" means "California"

http://amaexpo.com

Patrick talks to Matthew Schroyer from http://www.dronejournalism.org as per the picture above questions to the show @sUASNews and the #droneJchat hashtag on twitter

http://www.blogtalkradio.com/suasnews/2012/12/06/drone-journalism

Should start in just over an hour, but I am often rubbish working out the time differences, its should show you the correct time for your region on the blog talk radio link.

Just saw the new phantom RTF quad from DJI, cool design and incredible price (around $700 pre order). I think this can change the quad market ...........

http://www.dji-innovations.com/products/phantom/features/

The team at UAV Challenge Outback Rescue have released a photo gallery from the 2012 competition, the gallery includes hundreds of photos from the 3 days of the 2012 competition and Team Galleries

http://www.uavchallenge.csiro.au/2012/photos/Home/Home.html

Enjoy

Mick

I'd like to introduce myself to the DIY drones community, and show some photos of what I have been building. 

I am building a 50cc gas powered drone, with 3m wingspan. It will be all aluminium and foam and fibreglass, with no wood components.

Above is the solidworks image of the start of the design. I plan on using a hot wire cutter to cut the foam wings, and to have two voids down the center for two aluminium spars to go the length of the wings. Then, I will cover the lot with a single sheet of fibreglass and epoxy to keep out the 2-stroke fuel from the engine, and to strengthen the foam. I am using medium density foam as used by house renderers. Has anyone done anything similar to this before? There are many foam cut planes out there, but it seems nobody makes large ones. I think that the aluminium spars down the length of the wings, and the fibreglass skin will be enough to keep things nice and strong. I'll also add some aluminium tubes to connect the two spars together at the tip of the wings, to stop any torsional forces.

I haven't designed the motor mount and fuselage yet, but I am thinking of a pusher propeller at the rear of the wing, mounted to an aluminium plate which also secures to the wing spars.

The fuselage will have to be quite large, as I would like it to house:

1. Mini ATX motherboard (to run labview video recognition)

2. 10 Ah or more of 11.1V lithium batteries

3. Sony PTZ cam with 25X zoom

4. Ubiquiti Nanostation with directional antenna and associated tracking servos

5. Enough 2-stroke fuel for at least an hour long flight

I picked up this Sony camera on Ebay a while ago. The zoom is amazing on this thing!

As I have just had a baby, my man room has vanished. So I thought it would be a good idea to convert one of my trailers into a small work-room and storage area. And while I was at it, I thought it would be cool to deck it out with a multi-monitor permanent setup for FPV, flight sim, and GCS all in one. Then I could just hitch it to the car and take it to an air field. This would make setting up and tinkering a breeze.

Below is the solidworks model. I picked up a whole heap of Samsung 22" monitors cheap, and I'm running 4 of them off a matrox quad card, and the fifth off a USB to DVI connector.

I know that DIYD basically specialise in trying to get everything onto the one screen, but since I have the luxury of many screens I am looking for suggestions of GCS software which works well with a multi-screen setup.

Otherwise I was thinking one screen would be HappyKillmore's GCS, and the others would be google earth, weather info, live video feed, and command line telemetry respectively.

The box on the bottom left is space for an external petrol generator. I have a Honda generator which I am converting to run 2 x 12V alternators, which will charge 2 x car batteries. I have a 600W inverter to step up the power to 240V. This will act as a sort of UPS if I don't want to run the generator all the time, or if the air field has power, I can then just plug it in to charge.

The whole trailer will be insulated, and I will have an air vent directly over my head to push in fresh air. In Australia, the summer sun gets very hot!! Also, not pictured is an aluminium tube for the antenna stack.

So far, I have purchased:

Camera

50cc engine

Ubiquiti nanostation loco pair

20ah of 11.1V  laptop batteries

Spektrum DX6i

Xtend 900mhz module

All the aluminium

M3-ATX 12V power supply

Epoxy

Fibreglass

High Torque servos

Dome for camera

and a whole heap of other stuff......

Enough of the computer generated stuff.... Now, for the real build. I started building the GCS trailer yesterday. Here are some real life progress photos!

Please leave any suggestions! I haven't ordered the APM2.5 yet. I want to build the trailer and plane and make sure it all flies nicely. By that time, we will be up to APM3, or 4 or 5.... I do have an old ardupilot board that I will connect up to do some preliminary tests for the GCS tho.

Also, if there is anyone in Melbourne VIC Australia that flies with APM, please let me know which field you fly at, and I'd love to come down and have a good chat..... My local field doesn't really have anyone that does any sort of "technical flying" no FPV or anything fancy.

In this Post the manufacturing process of the Quadrotor used on the HyTAQ is illustrated. In case you have not seen the video of this system here is the link:  HyTAQ

Structural parts of this miniature quadrotor has been fabricated using the shape deposition manufacturing. To enhance the rigidity f the elements and also make the system more shock resistance, elements machined of aluminum sheets have been deposited in the parts. Also to eliminate the need to wires, the aluminum elements has been used to distribute power.

The first step was to machine the aluminum parts and wax block to the prepare the negative mold.

The aluminum parts are then placed in the wax block properly.

Liquid poly-urethane plastic with a shore hardness of 70D has been molded in the wax block.

The parts are then taken out and assembled.

IIT Rotocis Lab Website

First: This build can be followed in detail at: http://dzlsevilgeniuslair.blogspot.dk/

Long time ago I bought all the gear needed to build a small quad. I was really busy at the time so they ended up in a box and was quickly forgotten. The other day they reemerged and rather than just putting them away again i decided to quickly throw them together, at least they could sit on the wall and look cool.

I did not have any frame so i quickly lasered one out of plywood:

My intention was not to build an actually flying model, rather a fancy rack for the gear, so I made it quite small to save space on the wall. After mounting the gear I felt encouraged to take it a bit further so I actually connected everything and took it for at spin.

It actually worked pretty well so I thought that now it is a flying model the arms should be a bit longer and there should be some more features.

Really satisfied with the result both look and flying wice I coulden't stop there and made a camera pod

All in all I am surprices that what started as a quick and dirty jub turned out really well.

Testing a wing joiners. :)

You may have seen this guy's stunning UAV design work before. They were just renders initially, but now he's started 3D printing them, too. I wish I had more details, but his contact page says "Performing military service now. I will reply to the mail as soon as possible. I've asked nevertheless and will let you know if I hear back.

With the 3DR black friday deals, and the release of their power module, I took the opportunity to get another couple of APM2.5s, the cases that we have been anticipating, and some power modules.  I got it all assembled yesterday and I must admit I love the simplicity it adds to the build.

I now just have signal leads coming from the ESCs in to the APM directly. 

The 3DR power module sits in line before the distribution board, and that supplies the 5v for the APM to run from, as well as providing voltage+amperage readings from the battery.

With the case, and my QAV500 frame, I decided to use only the top portion of the case and mount through to the risers I already had. I also made a minor mod on the case for the UART pins, but everything else was smooth.  I cant wait to take it out for a flight.

The issue I was having with 2.8.1 not reading the amperage from the attopilot sensor is not an issue with the 3DR sensor (glad to see).

More photos can be found here: http://projectable.me/post/37261340554/yesterday-i-received-my-black-friday-specials-from

When I first got started with RC, one of my main learning resources was the excellent weekly podcast "All Things That Fly."  Two years ago they had some guy on with the crazy idea that the planes could fly themselves, and had a website that talked about open source hardware and software that could make such a thing happen called diydrones.com.

This week Chris reprises his appearance, talking about "drones, stabilization, geofencing, obstacle avoidance, 3D printing, CNC milling, and more."  Co-host (and commercially licensed pilot) Lucien Miller gives about the best summary of potential drone pilot licensing issues I've heard.

While you're there, be sure and subscribe to ATTF if you're not already a regular listener.  It should be interesting and informative to all diydrone.com regulars.

http://allthingsthatfly.com  (show #250, Dec. 4, 2012)

Hello all, I am a newbie from Christchurch, New Zealand (born in UK though). A few years ago (well, quite a few), I was very much into my Electronics & R/C - just for fun & challenge etc. I would often think that a drone could be a useful addition to the fire service, in order to help quench a raging forest fire etc! Now I see the UAV has really 'taken off'! I havn't been even close to R/C or my hobby, Electronics, for about 20 years due to my various disabilities,

But a drone could easily be programmed to refresh its water payload at a close lake or at sea etc. Many such drones could be utilised for a team of them to really help douse a fire? Technology has really made many quantum leaps & is getting faster & faster, so fast that we are dragging the chain in the ideas department to utilise this rapid growth. All's I mean by that, is often I see an outdated system or device that could easily benifit from modern aid.

It seems that unless there is a $ to be made, people are not that interested in helping other people that could really use some decent help from some decent people! I  can see some non combat UAVs being put to a good use (as seen in many videos) & so I think that you guys are the ones that can engineer such a system (a fire UAV) & do it for common good & technical challenge etc (enormous, but with technology & brains, some  AI  ...)! I guess you & your experiences in UAV, GPS, IR heat sensors and some very clever inginuity, may be able to flex your UAV muscles & make it all work!

I hope I havn't stood on anyones toes here, if I did, it was not intentional! I jusat hope spmeone may see the points that I am trying to make & share their interesting hobby & clever techniques with the rest of the world that is not aware of such efforts & achievments you are each doing, one way or another.

Thanks for allowing me to even get close to the real brains of the world! Enjoy your day, see ya!

Ricky.

[Image added by moderators. Source]

Here it is, the last thing we had to do before production.

Crash testing from Simon Jardine on Vimeo.

The model you see in the video is actually made of fibre glass and without a bottom plate. (it's not complete)

The end product will be made of Carbon Fibre.

Tests have shown the ring is more aerodynamic, has less vibration and overall feels more responsive and as you can see from the video its near indestructible ;-)

So far we have tested DJI, APM2.5, Multiwii,Mikrokopter,KK and the Naza all systems fit the frame and work well.

System specs:

Frame weight 200grams

Flight times of 10+ minutes

1240kv motors

8inch props

16.8volts LiPo 2700mah

Designed to carry the GoPro series Camera.

Production goes ahead 2013 January.

I loved RobotC back in my Lego Mindstorms days. Now it's available for Arduino, too. One of the big advantages of RobotC is one of the best debuggers in the industry, which is a glaring omission in regular Arduino (RobotC loads a special bootloader that tracks variables and provides that real-time debugging function). 

That said, I've tried this and couldn't get it to work with an Arduino Duemilanova, so I'm not ready to recommend it personally yet. 

From Hackaday:

There’s a reason we’ve seen a menagerie of quadcopters over the past few years – the key piece of any quadcopter build is an inertial measurement unit. Historically a very complicated and expensive piece of kit, these IMUs came down in price a few years back, allowing anyone with a few dollars in their pocket and a handful of brushless motors to build a four-bladed drone in their workshop.

[Starlino] built a few quadcopters, but he wanted to shy away from IMUs and get most of the mass of his new ‘copter over the center of the chassis. He came up with a design he calls the quadhybrid that can be built out of a quartet of those cheap 3-channel helicopter toys.

Most of the lift for [Starlino]‘s quadhybrid comes from a pair of coaxial rotors from a Syma 001 3-channel helicopter toy. Anyone who has played with one of these toy helicopters knows how stable they are; if the tail rotor breaks, you’re left with a helicopter that can only go up and down.

To give his quadhybrid a few degrees of freedom, he attached four tail rotors from 3-channel helis to a few booms laid out in a cross pattern. By taking the receiver out of a 4-channel helicopter and adding his own controller board, [Starlino] made each of the tail rotors control the pitch and roll of the craft.

In the video after the break, you can see the quadhybrid is amazingly stable even without an IMU and surprisingly agile. As [Starlino]‘s ‘copter can be made out of replacement parts for cheap 3-channel helis, we’ll expect a rush on these tail motors at your favorite online RC retailer very shortly.

Willow Garage's Robot Operating System was a great leap forward in creating an open platform for general purpose robotics (ArduCopter/Plane can serve as "ROS nodes"). The team has now posted a five-year status report, showing the growth of the platform.

From the post:

·      People often ask how many users are there of ROS.  Due to the open source nature of ROS, we simply don't know how many ROS users there are in the world. What we can tell you is that the ros.org wiki has had over 55,000 unique visitors in the last month.  This doesn't include traffic to our many worldwide mirrors.  

The latest version of ROS, Groovy Galapagos, is currently in Beta 1 Release.  Groovy will be the sixth full release of ROS.  This release is laying the foundations for enabling ROS to continue to grow the number of platforms supported.   

Inspired by The Mozilla Foundation, The Apache Software Foundation, and The GNOME Foundation, our three-year anniversary blog post discussed the possibility of a ROS Foundation.  In May of this year, Willow Garage announced the debut of the Open Source Robotics Foundation, Inc. OSRF is an independent non-profit organization founded by members of the global robotics community whose mission is to support the development, distribution, and adoption of open source software for use in robotics research, education, and product development.

Because of the BSD license for ROS, we often have no idea who is using ROS in their commercial deployments.  We suspect there are a few we are missing, but two major new products were announced this year that are built using ROS.  First is Baxter from Rethink Robotics.  Baxter was announced just a few months ago and the company has set their sites on manufacturing industries.  Check out IEEE Spectrum's article on Rethink here.  Also built on ROS is Toyota's Human Support Robot (HSR), which is designed to help those with limited mobility within the home.  ROS has even made inroads within the industrial robot world of late, specifically through the ROS-Industrial Consortium.

We can't discuss commercial deployments of ROS without mentioning TurtleBot, originally released in April 2011.  Recognizing that not everyone can afford, or even needs, a $280,000 PR2 robot, TurtleBot was brought to market for the express purpose of letting as many people as possible get their hands on ROS.  TurtleBot 2.0 was recently featured on Engadget and is now available for pre-order at www.turtlebot.com

At Willow Garage, we often refer to ourselves as a software company disguised as a robot company, and we can point to the ongoing growth of ROS as proof of that assertion.   We have also been stating for some time that we need a LAMP stack for robotics.  With the latest developments in commercial robots built on ROS, it feels like we are in the beginning stages of that process.  We can't predict what ROS will look like in five year, or twenty-five, but if we continue to see the adoption, innovation, and excitement from the ROS community that we have seen in the first five years, then things are certainly looking Rosey.

John Wiseman and Pat Hickey of the Mavelous team have been making great strides with this HTML-based MAVLink GCS, which can run in any browser.  You can see a very cool live demo of it here, and follow the project here. 3DR is supporting the project, so look for products from us soon that will allow you to use any device, including phones and tablets, as your GCS for ArduCopter/Plane in the field.

From Wired's Danger Room:

Four years ago, iRobot co-founder Helen Greiner stepped down from the company she helped turn into an all-important supplier of the military’s growing arsenal of ground robots. Now today, she’s unveiled the first ‘bots to roll off her new company’s assembly line. What are they? Teeny tiny hovering drones, designed to fly through your window and spy on you.

That’s just one of two robots revealed so far from Massachusetts company CyPhy Works, founded by Greiner after leaving iRobot. We’ve also now got a sense of what Greiner’s been developing for the past couple of years.

The first is Ease, or “Extreme Access System for Entry.” Really, it’s a tiny hover-bot designed for “intelligence, surveillance, and reconnaissance.” And it’s small enough — it only has a 1-foot diameter and a height of 16 inches from top to bottom – to fly through windows and maneuver through buildings with its ducted fan engine. In a video released by the company, the Ease can be seen hovering through an abandoned-looking building to a psychedelic funk soundtrack. It can also theoretically stay in the air forever.

“Being able to stay up aloft without constant interruptions to come down and recharge is a critical new capability,” Greiner tells Danger Room. “And with locations where you don’t have a lot of infrastructure.”

The reason is that instead of communicating wirelessly, the drone receives instructions and power through a microfilament cord of spooled copper the width of a fishing line and connected to the robot’s ground control station. And because it’s plugged in directly, the Ease drone should be harder to hack than other drones.

Once launched, a single battery at the ground station can also keep the drone up for 50 minutes, but this could be extended indefinitely by hot-swapping batteries at the control station. And as it is, most small drones that go wireless can’t stay up for very long, because “people have [started] putting more and more sensors and payloads onto them,” Greiner says, which drags on the power supply.

And while the operators are swapping out the batteries, Ease is scanning everything it sees with two high-def cameras and a third, albeit optional, thermal camera. Another purported advantage of the microfilament line is that the operator doesn’t have to worry about losing a wireless signal, say, if the robot moves behind a brick wall. And since the drone is moving in potentially close quarters to obstacles and people, “you really want to make it safe” by using a ducted fan, she says.

The other new drone is the Parc, or “Persistent Aerial Reconnaissance and Communications.” Like Ease, it also hovers. But the Parc is designed to fly high and for long periods of time, and resembles aflying bug with four skinny legs and a quadrotor. The robot can hover at 1,000 feet while being powered — like the Ease — by a microfilament line.

And while it’s up there, Parc can carry out “persistent stare capabilities” for up to 12 hours at a time (on one battery) while packing two cameras, one in high-def and another in thermal. And it has night vision and “flies itself.” (Oh great.) The military is reportedly interested.

There’s no word if CyPhy Works has received any orders yet, though. These drones are still prototypes. But it’s probably not far off. The Boston Globe‘s Scott Kirsner estimated the company’s investor funding at $3 million, “much of it from Cambridge-based [venture capital] firm General Catalyst.” But Kirsner also notes that the company has received millions more in federal grants. Which could make those tiny hover ‘bots begin flying into houses and spying on you from above sometime soon.

"Measuring Distance and Speed with a thumbtack-sized Radar Sensor

The SUCCESS consortium has developed and implemented a fully integrated 122 GHz Radar sensor. The surface-mountable package has a size of 8 mm x 8 mm. It includes a mixed-signal Silicon Germanium (SiGe) chip and two antennas for transmitting and receiving. The prototype is a breakthrough towards a miniaturized, low-cost and high-resolution millimeter-wave sensor for distance and speed measurements."

The press release indicates that the cost will be less than 1 euro.  No information on productization timeframe or effective distance.

http://www.success-project.eu/pages/news/press-release---september-7-2012.php

I decided to pick up one of those RCTimer APM boards a few weeks ago to help support free trade. This was before 3dR lowered the cost of the APM board. The kit complete comes with their clone of the APM, Radio Modem, and GPS. The kit arrived and I noticed many things are alike. Including the silk screening on the back of the radio modem that says "3d Robotics" that was covered up by the RCTimer sticker. Then again some items are different. One of which is the plug they give you for the radio modem. 

3d Robotics provides the opposite plugging of what RCTimer provides here. This creates a slight problem if you don't have a cable to swap things around. Thinking about the issue I recalled seeing UART silk screened on the bottom of the board. Flipping it over I noticed that UART 0 is labeled on one side.

Sure enough it's labeled UART0 just like the 3dR port you'd normally plug a radio modem into. After moving the wires around in the plug decided to give it a whirl and it works.

Thinking about it I realized something. Does this mean I can plug my regular 3dR modem into the normal UART0/2 port.. Then plug the Mini mOSD into the UART0 TX port on the side? Many folks talk about making a Y-Cable to attach the Mini mOSD AND Telemetry modems. Seems like adding a pin header on the other side and attaching there is cleaner.