Flying my Solo this evening when it flew off and crashed across the street after my son took a picture of it while it was hovering in front of him. No warning - just went. Solo was completely unresponsive to control attempts. See the event here: https://youtu.be/m368_A9hIwQ
Uploaded log to 3DR. Really lucky no one was hurt as kids had just been watching. Second flight - battery was around 20% - getting a picture just before going to land it.
Broke 3 blades but no damage to Solo or GoPro.
Doesn't build confidence when it just decides to do its own thing...
Replies
The question is not whether GPS can err - it's what the percentages are and what the best system is.
Articles such as:
http://electronicdesign.com/test-amp-measurement/real-world-drive-t...
written by the founder of the GPS antenna company, clearly spell out the advantages of modern dual systems.
"The tests were stark in their results: GPS/GLONASS dual-system antennas definitively offer a noticeable improvement in accuracy and performance. Urban environments are the true test of performance, and dual-system antennas are clearly superior"
In addition, these new machines are sold, marketed, advertised and claimed to be GPS quadcopters. That's the whole point! All of the advertised camera moves would be impossible without very accurate GPS.
Users should not blame themselves if a product is not engineered to the best of the modern standards....and I think that is the overriding question here. Perhaps the Solo GPS is blocked by it's being hidden by the large battery (from some of the sky)?
Whatever the reason, current technology allows better. Isn't that really the question we should be asking here? Should my $30 Moto-G outperform this model? It does - by a long way (without cellular).
In order to get the proper answers we have to ask the right questions. There is no doubt that every pilot should train, but that is not going to fix a basic hardware or other shortcoming.
How have quantified that it's better? (likely based on the fact that any GPS glitches in your moto don't make it fly into a wall ;-) )
The "without cellular" qualifier is an interesting addition to the statement If you you mean that the GPS performs better with the Cellular Radio turned off, that would also be true for a drone with GPS with it's RC Radio, WiFI radio and Video Radio* turned off. There would be less radio interference that would effect the GPS receiver. Transmitting antennas close to the GPS antenna is the main issue for GPS accuracy being disrupted.**
(I say GPS meaning GPS/GLONASS/BEIDOU they all have the same problem unless they are using some other medium than RF to transmit the data)
*If separate system is used like many people are using
**Mobile phones have the problem that GPS reception is adversely effected when you hold the phone
I never activated cellular on that device - and, as you know, cellular uses triangulation to improve the fix on our smart devices.
As to knowing that it's better - this is based on the fix being almost perfect using a GPS app - which also lists the number of sats, which ones it deems reliable, etc.
In theory lots of things could affect GPS. In reality, 100's of millions of these devices are navigating cars, boats and planes - even though the devices are inside a car (lots of motors and signals around) and on a small device running bluetooth, wifi, cellular and GPS.
The Real World seems an accurate place to measure the effectiveness. While I'm sure we can find examples of GPS gone bad, the percentages have to be tiny.
Some people are claiming multipath errors and reflections. This study by the maker of the GPS antennas - along with other similar studies - show an amazing ability of current dual systems to mitigate such problems...in the very worst environments.
http://electronicdesign.com/test-amp-measurement/real-world-drive-t...
The difference is stark.
Hi Craig,
I certainly agree that some of the newer modern GPS/GloNass chips are better than older ones and that 2 receivers can be an improvement over one.
However, that said relying on GPS for keeping your craft flying and safe is simply not reasonable.
There are too many circumstances where GPS cannot be relied on.
Direct signal blockage and multipath both continuously make GPS the least reliable system on your flight controller.
And that is with the receivers working perfectly.
And generally whatever outside influence is screwing up one GPS will happily also screw up the other one.
Now it is certainly true that you can detect in many cases that your GPS signals are not viable, but then you need an automated backup that does something like land or hover.
If you land you could be landing in the ocean or on top of something you don't want to land on.
And if you hover you are getting blown away.
Accels and Gyros can let you fight the wind for a bit, but not long.
Eventually we will have Optical flow and other ground "vision" referenced information that will give us superior automated options, but they are not ready for prime time yet.
Basically, the desired option is to retake manual control.
My thesis is that people need to be very comfortable with retaking manual control whenever - for whatever reason your automated control isn't going the way it ought to.
Of course in this instance, it very much looks like the person involved didn't have manual control in the first place, so for this use the point is moot.
I do think a company that is aggressively stressing the capability of its GPS automated functionality ought to be using at least 2 GPS and of the best capability.
Best Regards,
Gary
on both vehicles I built a dual m8n gps design has proven itself to be extremely reliable and works best with different types of GPS units, having both protected from EMI noise.
typical reason to have a crash and other issues before I went with dual gps were issues with gps cords, issues with batteries on gps units going bad, position errors and glitches.
dual gps unit on different chips does not glitch simultaneously, never. in a practical reality it means that you copter never looses its mind, tilts 45 deg and propels who knows where as it suddenly decided its new coordinates are not longer where it is.
dual GPS and dual compass - one internal and one in GPS is the only way to build those flying things.
combined then with position hold mode, properly working RTL, 'stop' and motors kill switch functions in new 3.3 rc7 it is finally a product that can be controlled in any circumstances.
manual modes are nice in direct line of sight and require quite a skill to recover a craft that suffers from malfunction.
Agreed - nothing wins like conservative flying (LOS), watching the wind conditions and being able to recover the bird by sight and manual (while still retaining baro and self-level) steering.
However - we are on the verge of a new era in these things and despite our roots, this is going to happen with or without us joining. It's no different than 35 years ago when you would not suggest the use of a computer to someone who didn't know how to use DOS, etc.
Aerial Robotics is the culmination of all the computing and mechanical technologies fusing and before long our hedges will be clipped by quadcopters. This is looking ahead rather than behind.
I went to a talk a few backs week - the Woz was just sitting on stage and answering questions and discussing life and technology in general. The guy is scary smart (of course!) and he did talk about camera drones - he has already flown the existing models. He said "I want one that can take pictures of me and other things without running into a tree, person or building"....meaning simply that AI and obstacle avoidance is the eventual goal of consumer (and many other) models.
We are probably only 3-5 years away from seeing some of these leaps.
As I mentioned above - it's not a matter of GPS being perfect or flawless, rather one of the percentages. Apple and others (every other new model of quadcopter!) didn't build in the GLONASS for marketing reasons - they did so because it is more reliable and accurate. What confuses me is that 3DR claims it is not. So either 3DR is wrong or every other tech and quadcopter company as well as the GPS antenna manufacturer is...
Hi Gary, 100% agree.
The real problem is that users trust too much the automatic flight.
I found this video on the web, imho these things hardly happen to an experienced pilot, and I'd be curious to understand what caused the drift of the Solo... an incorrect command of the pilot?
What i don't understand is why she often removes her fingers from the sticks, it's the wrong way to keep control in that situation, absolutely not to be imitated.
Not a very good place to do a selfie demo...
Hi Marco,
You are correct here. her logs were analyzed and was confirmed. She is an inexperienced pilot that was flying nose in and made corrections in the wrong direction.
An experienced pilot would not have been flying there at all GPS or not! That flight was irresponsible and ill conceived from the very second it took off in fact before she took off for even thinking flying there was a good idea.
There's simulator with Solo, is a good starting point for newbies... :p
What amazes me is that the girl seems to 3DR, i think i would be more careful about marketing.
"APM:Copter" is a fantastic software, but the user must know what he's doing.
-
1
-
2
-
3
-
4
-
5
of 12 Next