3DR Solo

[Gaza07]

Heres the first peek at the new 3dr solo it looks very nice with some excellent features and a pixhawk 2 controller 
who is ordering one and can I have a go  ::)

http://youtu.be/SP3Dgr9S4pM

[Hands0n]

I am really liking the new Solo from 3DR, it looks very polished.   At $1,399 it isn';t a terrible price, although i expect the GoPro is extra cost. So thats going to be knocking on for $1,800-$2,000. 

Given that 3DR recently open sourced their development kit for the mobile platform I expect to see many "applications" coming out for the Solo or anyone using an APM/Pixhawk with Tower mobile app. 

I had a play with Follow Me the other day.  The set up was an APM 2.6 with Arducopter 3.2.1, a Nexus 6 smartphone and the Tower app, a 433MHz telemetry groundstation module and an OTG cable (allows standard USB devices to connect to Android microUSB ports).    Running around the field like a demented escapee from the local institution, screaming "The drones are coming, the drones are coming" the hex followed me faitfully at the distance of the configured tether.  I was astonished at how very well it actually worked!

[powerlord]

Dji has sdk available too of course..  One of the things that attracts me to the phantom 3. Might even do an app.

Sent from my HTC One using Tapatalk

[JeremyE]

It is the best looking copter that they have released thus far. It is interesting that they state its got a pixhawk 2 and also mention running linux. Maybe its utilizing the pixhawk fire cape? It will be interesting to see when they actually release the next iteration of the pixhawk, I';d imagine it would only come once this has hit the shelves though.

[flybywire]

I wonder what market it';s aimed at?  Given that one particular criticism of the inspire is that it has 4 motors, so unlikely to attract commercial use?  Assuming of course that the powers that be outlaw quads for commercial work of course.   

Just being objective here.  At least it looks better than the last offering.

[Hands0n]

Here';s the full 6:44 Launch film

http://youtu.be/PfN9lTp8ma0


Exciting times seeing DJI and 3DR bringing out such quality kit.    And more significantly, open APIs for customer participation in the development of new ideas.

[Friskle]

I wonder what market it';s aimed at?  Given that one particular criticism of the inspire is that it has 4 motors, so unlikely to attract commercial use?  Assuming of course that the powers that be outlaw quads for commercial work of course.   

Just being objective here.  At least it looks better than the last offering.

Quite true, considering the slating the Inspire gets, and the mentioned possible restrictions on quads for commercial work, how will this fit in, is it being a quad such a good idea?  or more so, will it get slated ? or praised ? probably the latter, because it does not start with three Capital letters.

it looks nice though, and must admit, its features look very nice.

[quadfather]

Meh. 

Sure, it looks cool, much better than the rehashed Phantom 3, IMO, but I';m not inspired by the "launch movie".   The CG was cr@p and the playback choppy, as if it wasn';t optimised for YouTube.  It looked to me like borderline delusions of grandeur.  The "video game-style" controller looks too much like a DJI knock-off,  and the selling point seems that a monkey can fly it, as you don';t have to worry about the piloting because you can use a virtual tether.  What kind of message is that?

The Solo is available to pre-order for $999 at B&H or $1399 @ DSLRPros.  I wonder why the significant price difference, and if perhaps the latter comes with a GoPro?  Both stores sell primarily photographic equipment, so I reckon that the target audience is the aerial cinematographer, perhaps on the prosumer level (like the Phantom 3)?  The price tag would suggest this.  Both certainly outside my price range.

I don';t like the amount of proprietary components the Solo seems to have.  Can';t imagine that an additional battery will be cheap.  Do I have to use their TX or can I use my own radio?  I suppose I';ll have to wait to see the full specs.

Please don';t get me wrong.  I don';t hate it and and would gladly accept it as a gift, but at a first glance I';m not impressed.  Perhaps once the full specs are out I will be able to see what sets this apart from a Phantom 3.  I certainly like API features, and I';m sure a lot of cool stuff will from it (same goes for DJI).  It';s the aftermarket applications that will determine the real usefulness of this "drone".

[quadfather]

Perhaps once the full specs are out I will be able to see what sets this apart from a Phantom 3.

Well, ok, it has a PixHawk, of course!  In my defence I was in rant mode and didn';t feel particularly charitable.  :evil

[flybywire]

Well, bottom line with all these ';products';, is does it do what you need it to?  So far, a rock steady reliable fc with plug in osd, bluetooth, 2.4ghz downlink, hd lightbridge downlink, gps waypoint functionalty, plug and play compatibility with the best flight gimbals available is pretty much all i need, plus it all works perfectly, that, for me will take a heck of a lot of beating.

Ps, if you';ve ever owned a zenmuse gimbal, you';ll know what i mean!

[Hands0n]

I think it unfair to both 3DR and DJI to attempt comparisons between the two.  Fundamentally, I believe, they have different approaches to UAV outside of making flight controllers and aircraft that fly.   Although there is probably only so much you can actually do.  Comparisons are inevitable but detract.

3DRs focus has always been on automation, with even the now-humble APM1 pulling off stuff that nothing else available in the hobby world could.  APM2 built on that and Pixhawk has taken the capability to another level.  The development around Linux-based ArduPilot FCs will extend capability even further.  And I think that Pixhawk 2 is the first we are seeing of the pointier end of the technology in the FC.   Where is that all going? Ask Andrew Tridgell (Tridge) who seems to be a driving force in the developer community around Linux FCs. 

Solo? Definitely a prosumer UAS, and a good looking one at that. It is 3DR';s first ever complete package. I think we can all predict safely how this is going to fly. At least as good as any well set up Pixhawk-based quad possibly could.  The questions are going to be about what 3DR and the community are going to do with this bundle of technology.  I predict some serious robotics activities to come (within FAA/CAA regulatory frameworks).

[james]

The development around Linux-based ArduPilot FCs will extend capability even further.  And I think that Pixhawk 2 is the first we are seeing of the pointier end of the technology in the FC.   Where is that all going? Ask Andrew Tridgell (Tridge) who seems to be a driving force in the developer community around Linux FCs.

While I';ve not heard from Tridge or others on the Solo yet, I';m fairly confident in saying that the FC on this aircraft is almost certainly Linux based. They quote the processor speed as 1GHz, which is the same ballpark as the Beaglebone Black which has been used for quite a lot of the Linux FC prototyping. I';d be surprised if it were actually a BBB+Pixhawk cape in there, much more likely they';ve integrated it all on a board ala Pixhawk';s current incarnation.

Quite right in saying that this isn';t comparable to DJI';s offerings from a software perspective; 3DR';s kit has always exceeded what DJI';s can do, and this looks to be a step change once again from the older STM32 boards. Long gone are the days of really limited sensor filtering/fusion capability!

Competing for the Phantom 3';s market? It really isn';t. It';s the prosumer DSLR to DJI';s point-and-shoot rigs. If anything it';s competing with the Inspire, but that';s not a fair comparison either given the limited software capability of the Inspire.

[kilby]

I think that I would prefer the Solo simply because I have flown APM and have a Pixhawk that I still have to put into the air.

Unlike dji if there is an issue 3dr are seen to do something about it and I know the Iris was pretty well thought of (but to expensive for me)

Also I suspect the Solo will have a longer life as 3d4lr keep on adding value to their kit. DJI tend to smother their older products every 18 months or so

There';s nothing actually wrong with dji I simply have experienced and liked the competitors products.

[xxpitt]

I really like it. The tx looks awesome as well. I would have this over a phantom any day.  think I might like l the inspire a little more though

[flybywire]

Also I suspect the Solo will have a longer life as 3d4lr keep on adding value to their kit. DJI tend to smother their older products every 18 months or so

That';s not really true, as i mentioned earlier, they have added all those periferals to both the naza and wkm over a few years, plus the lite version.
The only thing they';ve discontinued is the v1, which was sperceded by the v2 of course, but v1 can still use the canbus etc, so i don';t see your logic here?

[Hands0n]

While I';ve not heard from Tridge or others on the Solo yet, I';m fairly confident in saying that the FC on this aircraft is almost certainly Linux based. They quote the processor speed as 1GHz, which is the same ballpark as the Beaglebone Black which has been used for quite a lot of the Linux FC prototyping. I';d be surprised if it were actually a BBB+Pixhawk cape in there, much more likely they';ve integrated it all on a board ala Pixhawk';s current incarnation.

Very exciting stuff if it really is their first production Linux FC.  I was impressed with the presentation by Tridge towards the end of 2014 where he demonstrated a wing running a Linux FC doing a ton of intensive internal processing (compiling the kernel if I recall correctly) while also being flown remotely using Auto mission and it did the circuit perfectly and landed effortlessly.   

At the time I recall Tridge saying that a multi-core Linux FC would probably use a single core for the FC and the other three or seven (the tech is changing rapidly) would be used for other processes such as in-flight real time image processing.

Of course, with a capable RTOS and multithreaded implementation of ArduPilot/Copter the capabilities of the FC software become almost limitless for all reasonable purposes. 

And if they';ve built in the various real time redundancies as in the original Pixhawk - fallback processors,vital  sensors and the like - the Pixhawk 2 in the Solo may just be a massive beast in power and processing terms.

I';m intrigued now.     

[hoverfly]

Now ime really confused i don';t own a space suit, monkey , or feel the need to swing flaming jam jars around my head.
Anybody got an idea what the new quad is like.

[Hands0n]

Now ime really confused i don';t own a space suit, monkey , or feel the need to swing flaming jam jars around my head.
Anybody got an idea what the new quad is like.

Leave it to the ad men to cook up something 2001 Space Odyssey(ish). 

We';ll get to read what its like when people get their hands on the thing.  Its not like DJI';s Phantom 3 launch where everyone know what a Phantom does anyway.    My view, however, is that this will be like any Pixhawk-equipped multirotor coupled with the Tower app.  Just made into an RTF package rather than a DIY kit of parts as currently is.

[HAUNTEDIMAGES]

i';ll wait till the pixhawk 2 comes out and then theres the clones etc.i';d rather upgrade current flight controller and use one my existing set ups.

[james]

Very exciting stuff if it really is their first production Linux FC.  I was impressed with the presentation by Tridge towards the end of 2014 where he demonstrated a wing running a Linux FC doing a ton of intensive internal processing (compiling the kernel if I recall correctly) while also being flown remotely using Auto mission and it did the circuit perfectly and landed effortlessly.

Ah not quite.

"No, Pixhawk 2 is based on the same Cortex M4 family as the original Pixhawk (but with new sensors, internal vibration damping, new connectors and more redundancy). The companion computer in Solo is a 1Ghz i.MX6"

The i.MX6 is not a slow chip, and personally I';m still quite happy with a coprocessor approach given that there';s not been much flytesting of Linux APs yet (Erle Robotics remain the only people shipping a real Linux FC at this time, afaik).

https://www.freescale.com/webapp/sps/site/taxonomy.jsp?code=IMX6X_SERIES&cof=0&am=0

The specific chip version sounds like it';s probably this guy:
https://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=i.MX6S

Makes sense given it has both CAN, a H.264 encoder (for the video downlink, I imagine), etc. They';ve confirmed it';ll all be open source hardware as well as software, so we';ll see schematics and BOMs soon no doubt for the hardware.

[teslahed]

I';m more interested in the Pixhawk 2 and want to know when it will be available separately and what advantages it really offers.

[Hands0n]

Interesting comment from the horse';s mouth (Chris Anderson, 3DR)

Comment by Chris Anderson yesterday
@Tsahi: Pixhawk 2 will be open source, just like Pixhawk 1. It is a derivative of ArduCopter, and will be merged back into Master. Most of the other software behind Solo is also open source, including DroneKit.

It isn';t loaded with specifics, but on the basis that "Pixhawk 2" is the entirety of the FC then the co-processor must reasonably be included in the OS definition.   

From all that I';ve seen of 3DR';s leading edge development intentions they have a focus on in-flight image processing.  Not sure what the practical use would be, but it is going to be much broader than simply beaming down HD video to the groundstation.  Although the ability to do so is not insignificant.  Image processing opens up a whole area of UAS yet to be truly explored. I';m thinking autonomous search and rescue, 3D ground mapping, agricultural and geological use.  Almost certainly use for hazard detection and avoidance using regular cheap optical cameras.  All stuff that can be done with an array of equipment today, but the potential to do on-board and in-flight where 3DR are going, maybe. 

If 3DR are true to form, the Pixhawk 2 is simply the hardware built and sold at an affordable price.  The "magic" is going to come from the developer community, particularly from the API availability.   At its most fundamental, Pixawk 2 I anticipate to be a FC at least as capable as Pixhawk (1) with more current sensors, so perhaps better flight characteristics.  All in what looks like a smaller package than Pixhawk (1).

[Hands0n]

I also anticipate some cool stuff to come out of their hook up with Qualcomm.

Sent from my HTC One_M8 using Tapatalk