And the biggest problem with the DJI products is that you have no real way to determine what went wrong. No tlogs or dataflash logs for post flight analysis.
Personally I have the suspicion that some of these indicidents are due to low battery. Remote location, test the hell out of the copter to make sure it works ok, draining the packs they have, don't have enough batteries and fly with a part depleted pack, for too long for the final shoot. And remember that last shot, he was in position waiting for the shot... flat pack could easily occur. And not really going to admit that on camera.
No, it was not better. I don't think any GPS glitch could affect our AHRS estimate that much. I've never seen it.
This issue with the DJI products is not just a simple GPS glitch, IMO. It seems to be a really bad bug in the firmware. I've seen several accounts of very experienced pilots getting hit with it. It's not just a case that people don't know how to fly in manual, but that the machine does not allow any control. You can flip to manual, but it just keeps zipping off.
Even Trappy got hit with this.
People often say "Stupid newbs, they should just switch to manual but they don't know how."
That's just an elitist comment from somebody who doesn't really understand the problem. I'm not saying all of these events are cases where the system doesn't respond to a mode change, but at least some of them are like that.
If it could have been a GPS glitch, there is one thing I don't understand. Why are flight control algorithms made with so much GPS and/or authority that a fail there can cause them to plummet into the ground?
I know an old-school system where the AHRS and the stabilisation/leveling worked without GPS at all and with minimal compass dependency. Flying autopilot, you could limit the maximum virtual stick input the GPS based autopilot had; with reasonable limits the copter would fly fast enough in auto, but would never zip into the ground.
Okay, the AHRS was not corrected for centripetal forces by GPS. If you wanted to fly the level estimate off level, could could do it (fly some tight circles). But who would want to do that?
Comments
Hi, there have been other controllers than DJI crashing on a GPS glitch too. Really shouldn't ever happen.
And the biggest problem with the DJI products is that you have no real way to determine what went wrong. No tlogs or dataflash logs for post flight analysis.
Personally I have the suspicion that some of these indicidents are due to low battery. Remote location, test the hell out of the copter to make sure it works ok, draining the packs they have, don't have enough batteries and fly with a part depleted pack, for too long for the final shoot. And remember that last shot, he was in position waiting for the shot... flat pack could easily occur. And not really going to admit that on camera.
No, it was not better. I don't think any GPS glitch could affect our AHRS estimate that much. I've never seen it.
This issue with the DJI products is not just a simple GPS glitch, IMO. It seems to be a really bad bug in the firmware. I've seen several accounts of very experienced pilots getting hit with it. It's not just a case that people don't know how to fly in manual, but that the machine does not allow any control. You can flip to manual, but it just keeps zipping off.
Even Trappy got hit with this.
People often say "Stupid newbs, they should just switch to manual but they don't know how."
That's just an elitist comment from somebody who doesn't really understand the problem. I'm not saying all of these events are cases where the system doesn't respond to a mode change, but at least some of them are like that.
If it could have been a GPS glitch, there is one thing I don't understand. Why are flight control algorithms made with so much GPS and/or authority that a fail there can cause them to plummet into the ground?
I know an old-school system where the AHRS and the stabilisation/leveling worked without GPS at all and with minimal compass dependency. Flying autopilot, you could limit the maximum virtual stick input the GPS based autopilot had; with reasonable limits the copter would fly fast enough in auto, but would never zip into the ground.
Okay, the AHRS was not corrected for centripetal forces by GPS. If you wanted to fly the level estimate off level, could could do it (fly some tight circles). But who would want to do that?
I wonder if old school was not better after all?
Regards
Soren
A must see for ALL commercial operators as this will most likely happen to you!
I suspect that compass is getting crazy by odd magnetic field in such a place.
"This thing is moving at whatever speed it is, with these carbon fibre blades."
I don't really see what the relevance of the last part is!
You have got to love dramatic TV moments
Yes, I'm with mdisher, it looks like a GPS Glitch to me. I wonder how well AC3.1's glitch protection will work.
The dramatization is hilarious.
I love the Youtube comments... "maybe they should find a better pilot" maybe it's too cold, "i bet they didn't calibrate".... etc...
The truth is they are cheap fun copters, using a DJI product for commercial purpose is wrong.
Like it or not :-)
Merry Christmas !!
Simon
-
2
-
3
-
4
-
5
-
6
of 6 Next