Today I lost my quadWent out flying and was hover about 10ft high and about 30ft away. I decided to try RTL turned on the switch and the quadcopter quickly started to climb in the sky above the cloud and disappeared.I lost my new APM2.5+ and Gps.(30days old) I am so hurt losing my new Apm2.5+I hope someone will fell sorry for me. If not I guess 1 more year of savings for a new one.
I always send mine off with a cheap GPS tracker unit from Ebay. The battery on it lasts around 3 hours from the moment you switch it on. I have found my Quadcopter a couple of times with this unit. Accuracy is around 10 meters. Use with any mobile phone to get the GPS co-ords.
very similar to the hype. could not return stabilization? out of this mode copter cant be armed before fly
but in general the telemetry may be way for make theft if use a powerful transmitter with a directional antenna by reconfiguration all modes to auto and using functions "change mode" and "go here"
"We are playing with DIY experimental semi-autonomous flying robots - a statement that I surely did not expect to be able to write in my lifetime when I was a kid ... "
I have read all the posts to this thread and however its nice that someone decided to give this gentlemen a new APM... I think it is important to learn the lesson here.
1) RTM/Wiki/Forums
2) If you are not sure, ASK.
3) If there is doubt, then RESEARCH.
3) These are not cheap toys, if you can't afford to damage/lose it, DON'T FLY IT.
God forbid that the Heli didn't land on anyone, or injured a child, adult or even damaged someone's property. You/We need to be responsible for the use of these UAV's. I am newbie, I don't even have my copter up and running yet. I am patiently reading, testing and reading and testing again. I am not dropping a $1000 into a Hobby to just have it run away and perhaps really do some damage.
IMO, if you can't come up with some reasons as to what went wrong. Then you clearly don't understand enough about what your are doing.
The often used cliche is "It's only fun till someone loses and eye!". Well, when something bad really happens, we all are going to suffer.
Don't get me wrong, I am not saying or speaking on your behalf.... I don't know you or how much work you put into learning to fly. This is more addressed to anyone thinking about getting into this.
Double check, triple check and check again before you do anything. These are not toys. I would bet that more than 75% of the flyers don't even have proper insurance to fly R/C.
That being said, I am sure I am going to get an ear full from readers... just be responsible and as a parent/grandfather, it would be no different than giving your children the keys to your vehicle when they turn 18 yrs of age.
@oliver - I completely agree with your approach. I just don't think we can count on everybody following a methodical approach. Therefore some default limits and sanity checks seem prudent.
@tim - I think RTL is a good fail safe as long as you do sanity checks by default before it goes active. Failing those, it would just autoland. Both those fail safes seem like they would be better to do most of the time in a radio failure situation than the APM default, which is nothing. All assuming you have a GPS. I haven't personally tried either because I'm not yet comfortable with stab, but if my radio were to go haywire, I'd appreciate the safety net I'm sure.
I think I found my answer to geofence...there's an allusion on the 2.9 thread that basically no one uses it and therefore it's not well tested. So now we have a catch 22....enable it as a fail safe and potentially risk a bug, or don't enable and possibly have a different bug or pilot error result in a flyaway.
By the way, for the record, none of this is intended as ragging on the devs. They clearly do way more than should be expected of a volunteer. And I'm sure if I were in their shoes, I'd rather be working on cool inertial nav features and such with the time I had to devote to the code and hobby. But from an end user noob perspective who's excited to get out and start flying, I think some attention has to be focused on a few of these underlying PICNIC errors that lead to catastrophic failures.
On the LED issue, I just got some super bright Chinese LEDs in the mail and I have some 5050 LED strips on the way. I also ordered a jdrones IOboard (they offer free shipping on light items to the US!) - I'm hoping I can program some of these desired light patterns using that.
About the climb out of sight: Did you check the altitude reported by telemetry before starting? I have experienced (every time, actually) that APM gets it initial altitude way off (10s to 100s of meters). This will cause it to believe the it is much higher/lower than it really is, and can explain this kind of things.
(What really happens: APM uses both GPS and barometer altitude and even an average of the two. The barometer must know the altitude of where it is initialized (ground) in order to calculate the pressure into an altitude. This offset altitude is had from GPS at startup. In my case, APM accepts GPS altitude much too early, when it is still very uncertain. Even when GPS accuracy later improves, the bad data remains in the baro offset where it was initially stored, and affects the baro and averate altitudes for the rest of the flight. I have to reset APM several times at start, until the altimeter showes the right ground altitude)
Simple solution #1: program a failsafe for low throttle descent... when radio is out of range, down it comes (hopefully slowly).
Simple solution #2: (the all in ballsy way) ...switch to stabilise, throttle to zero... assuming you're flying safely in clear land (you really shouldn't be flying over houses or populated areas, so this is a reasonable assumption)...
... wait...
... move to the sound of the "crunch/bang/boom" and hope you haven't started a grass fire with that lipo...
Collect pieces and pray that some are still useful.
So, what would you rather have... that it falls out of the sky when it runs out of batteries, or flies into a house/car/person... starts a fire where no is around to put it out... or simply that maybe having lots of broken pieces is better than having no pieces at all.
Replies
so sorry for you
Bad luck Marvin.
I always send mine off with a cheap GPS tracker unit from Ebay. The battery on it lasts around 3 hours from the moment you switch it on. I have found my Quadcopter a couple of times with this unit. Accuracy is around 10 meters. Use with any mobile phone to get the GPS co-ords.
I found your APM ... it drop in my head here in Brasil jajajajaj
Sorry Marvin .... but if you are not a good DIYer ... order a NAZA AP.
i am sorry but...
very similar to the hype.
could not return stabilization?
out of this mode copter cant be armed before fly
but in general the telemetry may be way for make theft
if use a powerful transmitter with a directional antenna
by reconfiguration all modes to auto and using functions "change mode" and "go here"
all radio modules going with same configuration
One ardu 2.4 and one older one (the one with a sensor-shield) is in the mail. Hope you will have fun with it.
Cheers!
"We are playing with DIY experimental semi-autonomous flying robots - a statement that I surely did not expect to be able to write in my lifetime when I was a kid ... "
wow, this is it. thank you.
I have read all the posts to this thread and however its nice that someone decided to give this gentlemen a new APM... I think it is important to learn the lesson here.
1) RTM/Wiki/Forums
2) If you are not sure, ASK.
3) If there is doubt, then RESEARCH.
3) These are not cheap toys, if you can't afford to damage/lose it, DON'T FLY IT.
God forbid that the Heli didn't land on anyone, or injured a child, adult or even damaged someone's property. You/We need to be responsible for the use of these UAV's. I am newbie, I don't even have my copter up and running yet. I am patiently reading, testing and reading and testing again. I am not dropping a $1000 into a Hobby to just have it run away and perhaps really do some damage.
IMO, if you can't come up with some reasons as to what went wrong. Then you clearly don't understand enough about what your are doing.
The often used cliche is "It's only fun till someone loses and eye!". Well, when something bad really happens, we all are going to suffer.
Don't get me wrong, I am not saying or speaking on your behalf.... I don't know you or how much work you put into learning to fly. This is more addressed to anyone thinking about getting into this.
Double check, triple check and check again before you do anything. These are not toys. I would bet that more than 75% of the flyers don't even have proper insurance to fly R/C.
That being said, I am sure I am going to get an ear full from readers... just be responsible and as a parent/grandfather, it would be no different than giving your children the keys to your vehicle when they turn 18 yrs of age.
All the best!
@tim - I think RTL is a good fail safe as long as you do sanity checks by default before it goes active. Failing those, it would just autoland. Both those fail safes seem like they would be better to do most of the time in a radio failure situation than the APM default, which is nothing. All assuming you have a GPS. I haven't personally tried either because I'm not yet comfortable with stab, but if my radio were to go haywire, I'd appreciate the safety net I'm sure.
I think I found my answer to geofence...there's an allusion on the 2.9 thread that basically no one uses it and therefore it's not well tested. So now we have a catch 22....enable it as a fail safe and potentially risk a bug, or don't enable and possibly have a different bug or pilot error result in a flyaway.
By the way, for the record, none of this is intended as ragging on the devs. They clearly do way more than should be expected of a volunteer. And I'm sure if I were in their shoes, I'd rather be working on cool inertial nav features and such with the time I had to devote to the code and hobby. But from an end user noob perspective who's excited to get out and start flying, I think some attention has to be focused on a few of these underlying PICNIC errors that lead to catastrophic failures.
On the LED issue, I just got some super bright Chinese LEDs in the mail and I have some 5050 LED strips on the way. I also ordered a jdrones IOboard (they offer free shipping on light items to the US!) - I'm hoping I can program some of these desired light patterns using that.
Hi Marvin,
About the climb out of sight: Did you check the altitude reported by telemetry before starting? I have experienced (every time, actually) that APM gets it initial altitude way off (10s to 100s of meters). This will cause it to believe the it is much higher/lower than it really is, and can explain this kind of things.
(What really happens: APM uses both GPS and barometer altitude and even an average of the two. The barometer must know the altitude of where it is initialized (ground) in order to calculate the pressure into an altitude. This offset altitude is had from GPS at startup. In my case, APM accepts GPS altitude much too early, when it is still very uncertain. Even when GPS accuracy later improves, the bad data remains in the baro offset where it was initially stored, and affects the baro and averate altitudes for the rest of the flight. I have to reset APM several times at start, until the altimeter showes the right ground altitude)
Regards
Soren
Simple solution #1: program a failsafe for low throttle descent... when radio is out of range, down it comes (hopefully slowly).
Simple solution #2: (the all in ballsy way) ...switch to stabilise, throttle to zero... assuming you're flying safely in clear land (you really shouldn't be flying over houses or populated areas, so this is a reasonable assumption)...
... wait...
... move to the sound of the "crunch/bang/boom" and hope you haven't started a grass fire with that lipo...
Collect pieces and pray that some are still useful.
So, what would you rather have... that it falls out of the sky when it runs out of batteries, or flies into a house/car/person... starts a fire where no is around to put it out... or simply that maybe having lots of broken pieces is better than having no pieces at all.
-
1
-
2
-
3
of 3 Next