Pixhawk or Naza?

If you are either building your own UAV or buying one (whether it’s an airplane, helicopter, or multirotor), the most basic part of the whole thing is the Flight Controller. Just like choosing windows or mac, iphone or android, choosing which flight controller you have in your aircraft changes the whole experience of flying. Stability, features, reliability, and customization options are different in each flight controller.
The two biggest platforms are NAZA (owned by DJI) and Pixhawk (previously maintained by 3D Robotics and now maintained by the Linux Foundation).
Which one should you use? If you’re a total beginner, use Naza. If you’re more than a total beginner, you might get more out of Pixhawk.
Below is an interesting conversation about it on Facebook. Take it for what it’s worth, but most of the experts seem to prefer Pixhawk, all things considered.


Dronecode Project – the LAMP stack of the 20-teens?

Exciting news in the depths of the engine room of the giant ship that is the Drone Craze of the 20-teens…. sUAS News breaks the story that the Linux Foundation in concert with many of the leading drone companies of today (3D Robotics, Baidu, Box, DroneDeploy, Intel, jDrones, Laser Navigation, Qualcomm, SkyWard, Squadrone System, Walkera and Yuneec.) are founding The Dronecode Project. (Yes, that’s everyone except DJI…)

It is worth pointing out that nearly every single drone startup that you’ve seen on kickstarter, as well as the great 3D Robotics company headed by Chris Anderson, use stuff from the Dronecode Project. It has until now been maintained by 3d Robotics in fact, and it is cool to see that they’re now handing it over to a neutral organization.

What is the Dronecode Project? Basically it’s a neutral governence structure for the major bits of open-source drone stuff that’s already out there. One of the major reasons that the drone scene has gotten so dramatically bigger in the last couple years is because of this open source stack of software. You might compare it to the LAMP (Linux, apache, mysql, php) software stack that was the technical underpinning of a huge portion of the Internet in the 90’s and 2000’s.

Dronecode includes the APM UAV software platform and associated code, which until now has been hosted by 3D Robotics, a world leader in advanced UAV autopilot and autonomous vehicle control.

The Dronecode project will also incorporate the partner PX4 project, led by Lorenz Meier from ETH, the Technical University of Zurich



“If you’re not crashing, you’re just not flying” – Richard Shelton


Richard Shelton of Multicopter Builders runs an online shop for drones. He designs and builds his own multicopters (as well as selling all the components you need to build your own).

His new quadcopter is called the MB Epic and it looks just absolutely awesome.

When it comes to the hot-rod world of high-performance multicopters, this man walks the walk. No empty promises or vague specifications here.

Do yourself a favor and watch a couple of these videos where the guy who builds and sells these copters shows what they can really do (oh yeah, watch this on a big screen!)


Plexidrone – another Crowdfunded Quadcopter – But is it realistic?

Kickstarter and Indiegogo are great for finding out where the hype is these days.

Drones are at the top of the hype curve it seems.

Indiegogo is hosting another quadcopter: the Plexidrone. It has raised more than $300,000 as of this writing.


While all the famous blogs are gushing over this thing (or at least, have made a very quotable quote or two), it bears looking into the specifications a bit more.  As with all pre-prototype, crowdfunded ideas, the wish-list is easy to make before the realities of economics and physics set in. As one friend of mine used to say jokingly, “No one can deliver what we can promise”.

I guess this quote, while at first blush sounds like a compliment, sounds a bit also like a warning: “If there was a list of features we wanted on a ready-to-fly drone, it might well read like PlexiDrone’s spec-sheet” – James Trew, Engadget

In other words, out of the hundreds of quadcopters that you can buy or build these days, if it was easy or possible to have all of these features, wouldn’t someone have done it already?

A couple of warning signals that I see:

1. the flight time – this is one of the golden specifications of any aircraft – how long can it fly? The Plexidrone people are claiming 35 minutes. Whoa! That’s a long time. The Phantom 1 flies for 5-10 minutes. The DJI Phantom 2 flies for 20-25 minutes.

If you look at the specifications of this very expensive motor ($200 per motor) that you can use as an after-market modification for a DJI phantom, you can see that – and this is a very expensive, very good motor – the ones in the plexidrone will definitely be worse than this –


So, for this “expensive ideal” motor, look at the Throttle and Amps columns. At 75% throttle, we are looking at around 5 amps of draw per motor. Let’s plug this into a handy flight time calculator (5000 mAh, 20A average amp draw) and we get 15 minutes of flight time.

The only way to get 35 minutes of flight time would be to use only 50% throttle with these (again, very expensive and not the type that is going to be used on the plexidrone) motors with 10 inch carbon fiber propellers – again, almost certainly not going to be used on the Plexidrone.

2. This Plexidrone weighs 1.3Kg; it has a 5000mAh battery – that’s around 410 grams. All the other parts therefore need to weigh 890 grams. The motors will weigh (at minimum) 100g each, so that’s 490 grams left over for everything else – the frame, the shell, the screws, electronics, and so on. This just does not seem realistic.


3. The flight controller. On the Indiegogo page, they state

What flight controller does the PlexiDrone use?
The PlexiDrone uses DreamQii’s custom electronics and software. We do not use any off-the-shelf options.”

Considering all of the features – including the dead giveaway – the “follow me” function – which is only available on Pixhawk controllers – it is obvious that the Plexidrone is using the Pixhawk controller. OK, Pixhawk is open-source. So they are using a Pixhawk clone. I guess it is still not flat-out lying to say that they are using “DreamQii’s custom electronics and software” but come on – it’s just a bit slightly un-truthy, isn’t it?

It’s hard to build a new flight controller – No one will think less of this company if they say that they have not built their own flight controller hardware/software from scratch.

It is hard to be one of the only naysayers around here willing to write a blog post that is not all rah-rah and spewing only positive thoughts…. but this quadcopter, as it is advertised, seems to be making some unrealistic claims. I hope these claims can be addressed and solved and that the people who back this project are not too unhappy when some of these specs slide downwards a bit back into reality.


TBS Gemini High-Performance Racing Hexacopter

It seems that for casual readers of technology blogs and sites that all these new quadcopters/multicopters/drones coming on the market are a brand new thing, often built by people who are new to the art of multirotors. This is often the case, perhaps because the veteran hobbyist folks who have been doing this stuff for years are in it for the love of the craft, more than the idea that there is some huge new business opportunity. So, what looks like a fantastic multirotor like this one probably won’t make it to Engadget, even though it is 1) available now (or, well, very soon) and 2) it is actually designed by people who know what they’re doing, with specifications that actually make sense.

The TBS Gemini is a fast and mean little hexacopter. It will go on sale for under $600, ready-to-fly without the transmitter. It’s perfect for beginner FPV folks.

These mini frames are very crash friendly. They are made to be flown hard and crashed. (Yes you need to be able to solder a few things back together from time to time)

You can read more about the TBS Gemini here.

Pre-order here for $599.

Again, this is not your Indiegogo vaporware… this is real, and definitely awesome.

Drone video showing the Hong Kong Protests (Sept 29)

A lot has been happening in Hong Kong in the last few days but it has been difficult to really imagine what things look like there. This video does a great job showing the tens of thousands of people that are filling the streets in Hong Kong.

The camera appears to be a Gopro. I’m not sure what kind of drone this is – there is a glimpse of the landing gear in a few spots but I don’t know what it is. At any rate, that doesn’t really matter – this video is a great piece of journalism.

There is a good discussion about it over on Hacker News.

Just in case this video disappears, I’ve made a few screenshots:


Drones on Kickstarter, 2014

a30b5503d8a159f4f252e6cb5c4125d9_large  Screen Shot 2014-09-30 at 11.23.37 AM  Screen Shot 2014-09-30 at 1.12.56 PM  0535897e177fe3e7a4b733b6ae21a73f

Multicopters have had a huge DIY community for years already, and while there are more and more RTF (ready to fly) slick-looking quadcopters you can buy yourself, the market is far from “entrenched”. The whole industry looks similar to what the automobile industry looked like 100 years ago: there are new companies emerging every day, each one with a slightly different take on the idea; many of these companies are true garage operations – and that’s a great thing, in my opinion.  (It’s worth nothing that, just as different brands of cars might use the same brand of tire, there are only a few of some of the components – the flight controller, most of all – that are available, no matter which drone you buy or back on Kickstarter)

Almost in tandem with the rapid rise of quadcopters  and DIY drones, we’ve seen the explosion of a “maker movement” around all kinds of hardware, as well as the rising popularity of crowdfuning sites like Kickstarter and Indiegogo.

So, let’s take a look at some of the newest multicopters out there.


The Phenox, built by some Japanese researchers, is the ultimate geek quadcopter. This thing looks amazing, although it’s more of a “developer’s kit” than a real consumer product. It is a (very) programmable UAV that doesn’t require an external controller. It can follow voice commands, or the sound of a whistle, or even hand signals. It is tiny but very good at self-stabilizing and following paths autonomously. This quadcopter has an on-board Linux computer (powered by a very powerful CPU / FPGA), motion sensors, two cameras, and a microphone. I think Ryo Konumura and Kensho Miyoshi have won the maximum number of geek points that can be earned with a project like this!

The video speaks for itself:




The Airdog is a quadcopter that follows you around. Presumably for the action-sports Gopro shooting folks. This looks like a solid team of young dudes who have crafted a very good startup pitch: They state the problem (When you’re doing that amazing 360 flip on your bicycle or snowboard, you need someone to be there all the time filming you), and then offer the solution (instead of having a friend following you around filming you, use the Airdog instead to follow you around from the air!). The way is works, basically, is that you wear a thing called an “Air Leash” on your wrist, and the Airdog follows it at a specified distance, angle, and altitude (all of which you can set on the fly) . Here is the Kickstarter video:

Successfully funded on July 26 2014 with more than $1.4 million, these guys have clocked up more than 1000 pre-orders (I mean “pledges ;). This is a textbook case of why Kickstarter is great for hardware startups (although I would guess they were fairly well-funded before their KS launch judging by the size of their team – even their domain name probably cost a fair bit of money!)

What is the Unique Selling Point of the Airdog? Well, they seem to be developing some very interesting technology around the “air leash”, allowing you to specify how this quadcopter follows the air leash. I don’t know what kind of technology they are using besides GPS for the tracking – GPS is only accurate to 10-20 meters, and if you’re going to frame a shot, well, I guess this might work ok as long as the drone is far enough away and you’re shooting with the very wide-angle lens of the gopro.

The biggest concern I would have with this thing is 1) its susceptibility to crashing into nearby things when it’s following you – buildings, trees, and so on; 2) how well it can land itself, and where it will land itself.

The Airdog is available for preorder on their website for $1300 with free shipping in December.

Here is their latest progress update from just a few days ago:




Easydrone could be seen as something in between the “absolutely ready to fly but you need to add the interesting extras yourself” of the DJI Phantom-type of quadcopter, and the “you want a gimbal and full FPV? You need to build it yourself” kind of quadcopter.

This drone has basically no wires – you can assemble it on the field, and the pieces snap into place, and some connectors are there to handle the various connections for motors, camera feed, and so on.  The central box contains the (open source Pixhawk) flight controller, FPV stuff, gps device, and the antennae.

I think the Easydrone looks very promising, although I would like more information about how the arms snap together so easily, but don’t fall apart easily.

Also, the design is clearly taking a second place over pure function. It looks more like a home-made multicopter than a Phantom, even though it is “plug and play”. The different models available on the Kickstarter page are not very clear.

It appears that the Easydrone folks are assuming a lot of technical knowedge for a product that they are selling as “ready to go”. This might at least partly explain why this campaign did not raise as much money as other ones (the other reason I’d say is the production quality of the video – this is absolutely correlated with the amount raised in a Kickstarter campaign… this Easydrone video got the message across, but, like the copter itself, it was a bit unglamorous). All together, it looks like the product might really be great, but it is not being marketed well enough – here is an example that shows me the importance of Marketing – from the conception of what the product looks like, to crafting a message that people can easily understand so that they can decide why your product is great. I hope the Easydrone folks can address this.

It looks like the Kickstarter campaign, which ended on July 11, is going fine. Its shipping date has slipped, which is really nothing too surprising (I don’t know any Kickstarter hardware project that actually delivered on time) and it looks like the updates are coming in fairly often.

If portability is a serious issue, and you hate wires, the Easydrone might be a good fit for you. If anyone has tried the Easydrone, please let me know!

Pocket Drone

Screen Shot 2014-09-30 at 11.24.40 AM

The Pocket Drone’s elevator pitch is this: “The world’s first multicopter that’s powerful enough to carry a high quality action camera and folds up smaller than a 7in tablet.” That’s good! In one sentence I can understand why this thing is cool. I would love to have a “real” drone (i.e. one that can carry a Gopro) that folds up to about the size of an iPad mini – how cool is that?

Screen Shot 2014-09-30 at 11.52.01 AM

Like the Airdog, the Pocket Drone has a “follow me mode” using your mobile phone (with GPS of course)

It also seems to allow control of the drone via an Android phone, but NOT using wifi, but rather using a separate radio transmitter dongle that you plug into your phone. This is very cool.

Their video is well-written and gets the message across that these guys are building this tricopter because they really want to have this great toy for themselves. They appear to be participating in the DIY multicopter community for a while, and have designed a product that really hasn’t existed up until now.

I guess the biggest caveat for me would be the fact that without a stabilizing gimbal, you can only film the “roller coaster” style of video – it does take a lot of practice to shoot that kind of video without making your viewers sick. Anyway, if that is the biggest problem with this thing, I would be more than happy to keep one stashed in my backpack for those many spare moments in life that I could be flying a copter and making new friends 🙂

The Pocket Drone had a smashing success on Kickstarter, raising nearly a million dollars and getting around 1,600 pre-orders at a ~$500 price point – this is fantastic to see. The product looks really viable to me and I hope that they manage to ramp up production and work out all the challenges they face.

The current price for the Pocket Drone on their website is $599 with controller. This is a hard sell considering a DJI Phantom 1 can be purchased for $479. If you like the portability of the Pocket Drone, then I could see why it would be ok to pay more for this obviously very cool new product.

As usual, it was supposed to ship in June but isn’t shipping yet, but it appears that they’re working really hard to get everyone’s pre-orders shipped soon.

Game of Drones

This is not really a whole drone per se – it’s just the airframe. It appears to be the most outrageously tough quadcopter frame in existence. Also, they have the coolest logo ever:

download (1)

These guys say their super-tough quadcopter frame has been crashed thousands of times, landed on water, flown through fire, and even shot at with a shotgun (!!) and it is still flying. These guys look like true multicopter maniacs who are clearly doing it for the passion of making something cool:

Their latest update says that they are finally shipping now. Good job guys! You can buy a frame right now at their shop for $140 each, or a 10-pack for $900.



This hexacopter is another “follow me” drone which is aimed at action/sports people who need a flying robot slave camera man to cover their crazy moves (are you feeling like a lazy slob yet? I am!). It can fly up to 70km/h for about 15 minutes (wow, that is a lot). Also, I don’t see a landing gear anywhere but presumably they’ll solve this!

Screen Shot 2014-09-30 at 1.45.06 PM

They have plenty of stunningly beautiful example shots in their video. I’m a bit skeptical of the shot following behind the runner, going between (and under) two rows of trees – I mean, maybe this thing can do that, but how many people are going to risk it? Does it really work that well? It is a pretty bold claim.

From what I can tell of some of the shots in their video which I presume are mostly mockups, I can guess that they are trying to do some computer vision / object tracking in addition to simply using GPS as a beacon (you would have to if you’re going to pull off the shot flying under the trees at 2:03….

This is another “fabulously successful” Kickstarter campaign, having snagged nearly 2000 pre-orders. From the updates being posted, it appears that they are seeking VC funding in addtion to the $1.3M they’ve raised already, which should help them ramp up their development and manufacturing.

The Hexo+ is still available for pre-order at $949 (with 2-axis gimbal) and $1149 (with 3-axis gimbal) with a shipping date of May 2015.


iStrike Shuttle

This one is actually from the end of 2012, but it is so wacky that I had to include it in this post.

I have actually never seen this form factor until now. Behold, the iStrike Shuttle:

Why, you may ask, is it shaped like this, with the upright, elongated body? Well, I’m glad you asked! This thing deserves to be called a drone – in a way that many “multicopter” pilots find the word objectionable – you see, it’s build to drop ping pong balls! That’s right, you can actually invade your neighbor or office mate by piloting this thing with your iPhone, dropping ping pong balls on them.

As this project ended on Nov. 23 2012, we can even find a bit of evidence that they did, in fact, fulfill their backers, and produce these things. Here is a random video from youtube, presumably from one of the backers. While the video is not long, it shows that this thing does, in fact, fly:

The iStrike Shuttle is now available for sale on the Dream Cheeky website and it is $59….


That’s all for now. If I have missed a product, if I’ve made any mistakes, or if you have any other question please contact me.

Gopro Hero 4 Released to the wild

Well, it’s finally here.

Photo by Jimmy Chin

4K resolution, 30FPS video. Full HD, 120FPS.

It is the same size as the Hero 3, so it will work in all our attachments and gimbals.

Still, no RAW photos, sadly. However, there is a “Protune” setting for jpegs, so that you can shoot photos that aren’t “so totally automatic” and over-compressed. Tests coming soon.

I guess all the other action cameras out there have fallen one more step behind. There is just no stopping Gopro it seems.

So, a brief rundown of what it means for drone pilots:

Yes, this is a drop-in replacement

It’s the same size so just switch it out with your poor old crusty obsolete Hero 3 that you bought last month 😉


Will the lens mount modification still work, or will it still be necessary?

Gopro cameras have been using a very strange kind of lens which is nominally C-mount (m12 thread) but with the bottom part of the lens not threaded at all. This meant that in order to use a diferent lens, you had to remove the entire lens mount and put in a new (3rd party) one. I am guessing that this is still the case. So the question is, will the previous ones work on the Hero 4? I am guessing they won’t. Probably the PCB has undergone some tiny revision, and a new 3rd-party lens mount will have to be devised.


Will anyone be able to actually watch my 4K video?

Well, maybe not this year 😉 And you will need a beast of a computer to do any real post-production with such files. Get ready to buy some spare hard disks to store all this footage….



3d Pocketcopter raises more than 100K EUR on Indiegogo

The first time I saw this thing…. I was intrigued and I seriously considered plumping down 79 Euros. But wait, I thought…. how is it supposed to steer? How big is it? What kind of camera? How can it shoot 3d using a single sensor? Huh?

It is worth nothing that Indiegogo does not require “a working prototype” as Kickstarter does. So this project should be taken with a grain of salt.

Since the project has closed, it appears that the creators have at least published a single technical drawing of how this helicopter might work. In short, it is a traditional helicopter – not a multicopter – the difference being that the large rotors can lean in one direction or another using a swashplate, in order to steer.

What I’m (sadly) most impressed about with this project is the evidence of how easy it is to raise more than 100,000 EUR without a prototype or even with an actual proposed design. The device is simultaneously complex and beyond the understanding of most people, but also sort of small and cute, and seems like it could be pretty simple…

I will be a bit surprised if this device actually makes it to fruition in this incarnation, form factor and price point, one can hope. Good luck to the 3d Pocketcopter people!


Nixie Drone bracelet (a half-baked prototype)

Ever since the word “selfie” entered mainstream in the last year or two, there is a small portion of the population which seems interested in bringing “selfie” to a new level by letting people operate their own personal drone in order to shoot movies of yourself from the sky. Enter the Nixie, which seems to be really only half-baked, but let’s take a look:


I am not at all sold on the idea of a “personal drone that follows you around”. Maybe it’s because the quadcopters / drones I’ve flown are rather susceptible to crashing into things. I wouldn’t call this an easy problem to solve either. Until these aircraft have very impressive abilities to detect obstacles…. including, most notably, other people, who probably don’t take kindly to getting hit in the face by a propeller, these things are going to have a bit of an uphill battle. I don’t want to be too pessimistic, but honestly I think these things are at least a few years off.

Drone made of Lego succeeds on Kickstarter


A gadget is not cool until someone makes it out of Lego… and then makes a Kickstarter campaign out of it 😉 So I’m happy to say that finally someone has done just that with a quadcopter. Behold, the Lego Brick Drone Frame:

This project is a true DIY / Maker kind of thing; you still need to buy your own motors, ESCs, flight controller, and battery. What you do get are all the necessary bricks, and 3d-printed motor mounts. What is pretty impressive is that this thing can carry a Gopro! I guess you probably want to glue those lego bricks together if you’re going to carry a $400 camera…

This project embodies what I love about the maker culture of today – I hope it works out for everyone involved.


FAA welcomes Drones for Hollywood

The NY Times reports that the FAA has finally started to approve some drone use for commercial applications, starting with six Hollywood film production companies. The companies are: Aerial Mob, Astraeus Aerial, HeliVideo Productions, Pictorvision, Snaproll Media and Vortex Aerial.

This is great news for the industry and a step in the right direction for the FAA, who until now has been dragging their feet, and letting the USA fall behind other countries in the commercialization of drone operations.