OK, THIS IS A RANT!!
I am done with Arducopter! I have been trying to get this platform to be reliable for way too long.
I don't even want to add up all the damage, time and money I have wasted trying to get this flight
control system to work as advertised.
I actually thought with the new firmware and M.P., that the developers had finally got out all of the
As I write this, I notice that the top discussion is unexpected scary start-up.
This is what happened to me:
I dumped all programs from my computer, and reset my APM 2.5, I started over.
I downloaded the new versions clean, without any of the zillions of updates.
I was so excited to see that my project was finally working very well, a doing what I told it to, and it
worked very well for two days. On the third day, without any changes at all to anything, it did a full
throttle cut. I turned it off, rebooted my computer, and started over. It behaved normally for about two
minutes, then two motors cut out. I was only up about six feet, and over a lawn, so no damage.
As I approached my machine, to unplug the battery, two motors started to spin at different speeds.
Then they all spun up to full throttle. I always carry my Aurora 9 with my left thumb on the throttle
stick, so there can not be an accident.
OFF IT WENT, into the sun. Hit return to home, no joy. I said goodbye to it, as it went to an unknown
altitude, and into the sun, (downwind, it was just trying to go straight up).
I found it the next day, see the picture. Two miles of walking and searching and cussing.
I went over the logs, and determined that there were error codes all over the list, and it breached the
geofence, but kept going anyway. It had a mind of it's own, TOTALLY NOT COOL, and VERY, VERY
If this had happened in a more densely populated area, I might be in jail right now, or being sued, or
worse. "Drone kills baby", on the eleven o'clock news. My machine is big, heavy and damn-near
indestructible. I do not know how far it fell, I going to guess at somewhere between 700 and 1000 feet,
(angle of sun, do the trig.) Now it's broken.
I want the FAA to let us do this, but if these kind of failures keep happening, someone's going to get
hurt, and then the Government will make it illegal! Game Over!
What is up with all the flaws in this platform? Ardu may kill any chance for guys like us to go out and
make money with this tech, before the FAA and Congress even make their decision next year.
Oh, and by the way, I was to show and demonstrate my machine to a government contractor, with a
C.O.E., who shall remain nameless, the very next day. I got to show them a hulk. The only plus was
they were impressed by my build, because it's still repairable, and the expensive stuff lived.
My suggestion to ARDU is to get your shit together. Something very bad is going to happen, it already
has to me, and if someone had gotten hurt the other day, and I was being sued, I would sue ARDU.
Then it's all over the news, and all over for us. This "mishap" was of no fault of mine. Does the
software rewrite itself? Did someone embed malware into it? Did the hardware just "decide" to melt
You guys need to perfect this, or it's not going to go well with the FAA.
If they were to ask me if it is a safe and reliable platform, I would have to answer "HELL NO!, take it of
the market, before somebody gets killed." I am not giving up on this technology, just Arducopter,
and SO SHOULD YOU ALL!
Replies are closed for this discussion.
Sunlight? Never happened before, but I guess that makes a little sense, but I was fling on a very cool day, about 45 degrees out, and a low angle of sun, so I don't think that was it. I have operated it just fine on hot days, in full sun, without trouble.
Is this what it looked like before the crash?
Is that the RC Timer dome? Who's GPS is that? Kinda almost looks like a Mediatek? I can't really make it out, but it looks like the APM isn't in a case. So is the baro covered?
You say it flew into the sun. That means the sun was East-South-East of you (morning?). So when were flying, and landed, it's back was to the sun. This is the angle the sun was hitting *my* quad with an uncovered baro when it had a problem. In my case, it dropped like a stone because it wasn't affected by the sun until it turned after the first waypoint. I tried rerunning the mission 5-6 times, and it always fell soon as it turned with it's back to the sun. The sun was to the south, I'd take off facing east, and soon as it turned north to waypoint 2, it dropped like a rock. Every time. I had to keep bailing out into Stab, and Randy pointed out my mistake (baro uncovered, we all make mistakes).
How are you feeding the 5V into the APM? As I said, the Vcc is absolutely terrible. Are the gimbal servos drawing from that same VR? Are they drawing *through* the APM?
Yep, exact same thing with me. I was flying... maybe it was late October, sun was getting low at about 2 o'clock. First time I ever had a problem. It definitely seems like the light has to come in at some angle. I also had an upper deck over the APM. The sunlight has to sneak under that.
It's just a guess, but I find it highly suspicious that the altitude estimate stayed below zero while you were landed and disarmed. Randy might know more about how that can happen.
So here's what I've got so far. Trying to line up the events with the Tlog playback. You definitely armed it, and lifted off in Stab mode, and it rotated a little bit. So if by my theory, you had landed with the sun on the baro, disarmed/rearmed, it would have saved a false off-set. Then you took off, sun came off, and the false off-set pushed the altitude way up, at which point it thought it had a fence breach. It seems to me that it actually landed WHILE IN RTL, and you didn't even know it. You dropped the throttle and thought you landed it. Unfortunate co-incidence. It was sitting on the ground with the motors running lightly right?
What you didn't realize was it was actually trying to RTL. I actually can't tell what happened after that. It's clear that it decided to throttle up and take off. I have no idea why this would have happened, as should have thought it was in the right location. Your RTL Alt is 2500, which I think is 25m, so that's fine. Why wasn't it satisfied with the landing? It might be because it thought it was too high still.
It kinda looks like it danced on the ground a couple times, rotated, sunlight maybe just coming in and out... Just a guess. But look, every time the yaw changes, the Alt makes a big change.
I know that having all your hard work destroyed, by something that seems out of your control is very upsetting. But the only benefit that comes out of it now is to look at the situation, figure out if you did something wrong so you can fix that. Or perhaps generate "lessons learned" for others. Or, it still could be a bug, or just something we could error check better. I can't make much more out of this, and we'll have to see if Randy or somebody else comes up with something.
I'm baffled, and I don't read code either, I'm a builder, a maker, at the Maker Faire twice with large helis. I saw 3DR's staff flying over people, in the wind there, not cool, and against AMA rules.
I fly 500's and larger and am very safety conscience. These anomalies can not be tolerated in a professional application, and are dangerous to everyone else. I've had throttle cuts before, spaz-outs, and flip-overs, but never an uncontrolled fly-away.
Yes, please have Randy check it out, I can't afford a Wookong right now and Aireware is even more.
I'm starting to get very afraid of liability issues, and so should you.
Peace, and thanks for getting on this.
Arducopter cannot be expected to protect people from themselves, especially when the users builds the machine themselves. If this was a standard Iris or something, you might have a point, but this is a machine built almost entirely by yourself, and YOU bear some responsibility for what happens with it. The people who write the code can only try to make it behave properly according to the information it's getting from the sensors. If the sensors are giving crap data, then what really do you expect the program to be able to do with that?
So the questions remain:
Why is the Vcc so bad? Are you driving those gimbal servos *through* the APM power system? I bet you are.
Is the baro covered? I think it may not be. Is it, or not?
How much vibration are the propellers putting out? Foam or not... you can't put an APM on foam, and attach it to a jackhammer, and expect it to perform. Have you ever checked the vibration levels with those particular props? As I said, I tested props just like those, and they were garbage. The problem is, they're fine at low power. The problem only occurred above 60% throttle, then they started to flutter badly. That fluttering would completely upset the INAV system.
All of these issues are covered very clearly in the wiki.
Now, I'm not trying to blame this all on you. I do not have an explanation for exactly what happened afterwards. I have a suspicion it could be an unfortunate series of events, involving both the sun on the baro, and vibration. I don't know for sure yet, we don't have enough data.
Speaking of data, how come you didn't know that it was in RTL? Do you have a ground control station?
My personal opinion, is nobody should ever fly with any automatic modes enabled without a GCS. The system is simply far too complicated for anybody to understand without feedback. If you had known that it was RTLing, and changed modes back to Stabilize, you would have likely regained control.
Moving forward the FAA has chosen its first test sites and the DHS has a plan. Anybody looking to do commercial work will have to get their platforms through the hoops or buy one that has been through them.
On the GCS side, I am coming to the opinion that perhaps the AP should not be able to fly unless its connected to a GCS as well as a standard RC controller.
This would all have been a huge pipe dream 5 years ago as tablets and smartphones were barely around or powerful enough back then. These days they are everywhere.
I only use my tablet for the voice announces, I know exactly what my platforms are up to in flight. Funnily enough the tablet is exactly how I go into auto modes as well, having loiter RTL and stabilize only one my RC controller. Every time I see 47 modes whilst standing on your head from your TX threads I shudder. Its making a simple thing complex. The other advantage of taking a tablet out as well is should your RC fail you have another emergency command and control link.
RC model flying and pure UAS flying are similar but also very different. There has to be a much more planned approach to what you are doing with a UA. Before it takes off you have to have a clear picture in your mind of where its going to go and how long its going to take.
As they say PPPPPPP
I actually don't even know how to switch modes from MP! :) Never tried that. I actually looked the other day and couldn't figure it out.
I would just like to say that I'm getting a little tired of all the folks out there who think that this technology is military grade. I go back to RC in the 1970s, where a good day consisted of three flights and bringing your plane home.
APM is unbelievable. I see more and more modelers, or folks who think they are modelers, believing that everything should be 100% foolproof. It's not. I got into this hobby as the thought of flying something by remote control was absolutely amazing and unbelievable. With APM I have seen those dreams come true. It still takes a little work, a Lot of reading, and ever present due diligence to make things work successfully. This is a hobby, even a sport for those who compete, and those flyers out there who expect everything to work as a seen on TV are naïve. Progress is made through the efforts of hobbyist who push the envelope, and suffer the consequences. I cannot even fathom the loss of the good times that I have had in pursuing these goals. Everyone in this hobby, and I mean hobby, must understand this. If your goal is to fly a fully autonomous vehicle, with no flight skills than perhaps you should look at pursuing a different sport. The work of the folks who have developed this software and continue to refine it cannot be understated. They have put many hours, and many wrecks under their belt in pursuit of developing software that allows we commoners to enjoy the fruits of their labor. Coming from one who has built many RC devices from scratch I can attest to the fact that the work needed is very time-consuming and often unfruitful. The ability to buy hardware and software that works is absolutely fantastic. If one simply wants a system that works go spend four or $5000 and buy something that is ready to go out of the box. But be prepared for it to fail as well as this is not technology that is fully developed. I for one work for every dollar I get. I don't enjoy a crash anymore than anyone else. I often find that most of the problems I've had are as a result of my error, not as a result from the work done by the talented folks who put their time into the development of this software. I read everything I can find on the Internet, and I am thankful for the Internet as when I started in RC the only way to learn anything was through conversations with other modelers at a hobby shop or at the local flying field. I understand hobby shops are dying breed and that everyone relies on the Internet, but I cannot believe that everything I read on the Internet is 100% true. It often reflects the experiences of that person at that time for that particular purpose.
I urge that everyone, like it or not, that if there's a problem with your equipment or your setup that you ask for help, and accept whatever is given; understanding that it may or may not be the solution to your particular problem. I guess this is going to be considered as my ranting and raving, but I have had far more many wonderful experiences flying my APM equipment that I have probably had in my entire lifetime.
Take care all.
Don't know but that Arducopter has a killswitch now? You can not have a machine that is supposed to be *somehow* autonom. without that option - just my opinion. Retired my apm for similar reasons.
What about changing/extending the has landed/isn't flying condition? If the total throttle is below a specified minimal fly throttle (maybe esc minimalvalue + 5%) for a certain amount of time (like 5 seconds) disarm - no matter what.
Maybe users should be advised in the wiki to throw a blanket over the still powered copter on the ground before approaching it.
Is there a dataflash log somewhere? I see the tlog but no dataflash...that would be helpful.
To bring something positive out of this, is there an option to have some form of barometer failsafe introduced into the code? Its unlikely someone will be flying at -190 unless they have the UAV inverted and are tunnelling for oil ;-)
I have always run my APMs in their cases with foam protecting the barometer and each of the UAVs have a dome that is custom made from dark smoked acrylic. Some could say its going overboard, but its all about negating risk (and it looks cool).
I am about to take my commercial license in the UK and as part of the process you have to submit a "risk assessment" document. These processes do make you appreciative to what "could" go wrong and make you think more about your use, but its equally about balance and doing everything you can to negate or reduce risk.
This APM and PixHawk technology goes through rigorous testing by some very talented engineers. The only people that are going to spoil it for the masses is the users. Its like handing over a loaded gun, we have to treat it with the utmost respect. A gun will only go bang if someone does something wrong and pulls the trigger, you can't blame the gun for that !