With both my Mavic Pro and Mavic Pro Platinum flying 6-mile round trip Litchi waypoint missions over the past couple of days, I have encountered an identical camera anomaly which in both instances occurred during the last few minutes of those waypoint missions launched out over tropical rainforest.

The first portion of these two test flights was uneventful, but then all of a sudden the camera began to rock violently in the yaw and roll directions, while simultaneously Litchi’s audio announcements repeatedly intoned the warnings “Camera Stopped/ Camera Started”, while the camera view horizon tilted wildly.

At the very start of all my Litchi waypoint flights, I generally tilt the camera down to a minus-20 degrees downward attitude to maximize the ground view, while leaving just a narrow sliver of the horizon in view at the very top of the screen. In these two instances when the camera of both my Mavics took on a life of its own and proceeded to flail about, the camera tilted upwards until the horizon and sky filled more than 60% of my iPad’s Litchi camera view screen, while also rocking left and right in a most alarming manner.

  Attempting to roll the camera back down to my preferred minus-20 degree tilt angle did not work while the camera danced about during those flights, because there was no response at all when I turned the wheel at the top left corner of the controller. The camera continued displaying abrupt and random movements most of the way home after I actuated RTH, with the effect tapering off a little as the drone approached within audible range.

My initial suspicion was that a simple SD memory card replacement might resolve the looney camera syndrome because I recently discovered that failure to re-format the SD card prior to each flight also disabled the camera-tilt thumb wheel, and disabled the record function. This is pure speculation on my part though, as is my conjecture that an SD card flaw is more probable than the odds that BOTH my Mavic Pro and Mavic Pro Platinum could develop identical in-flight camera failure symptoms within days of each other.

  This mystery bothers me enough to post its description in this and one other forum, just in case any Litchi and/or Mavic Pro users have encountered such sudden departures from normal camera behavior during waypoint flights or even during manual drone flights. Any ideas that can be shared would be gratefully considered here, as I try and figure out what might be amiss.
  

Hello there Muttley7. I am sorry for the trouble this has caused and thank you for reaching out. Since this is a 3rd party application ( Litchi ) that you are using, I hope that you will get the best recommendation from our fellow DJI co pilots who had used this application. Just a reminder that DJI does not recommend you to your 3rd party application when flying your DJI Drones. These 3rd party applications might affect the said DJI drone when in use. There is a possibility that these 3rd party applications will be able work. However, since it's not optimized and tested by DJI we cannot guaranteed the stability of the said 3rd party applications if it will work all the time with no issues. Thank you and have a safe flight always.

DJI Stephen Posted at 6-14 21:19
Hello there Muttley7. I am sorry for the trouble this has caused and thank you for reaching out. Since this is a 3rd party application ( Litchi ) that you are using, I hope that you will get the best recommendation from our fellow DJI co pilots who had used this application. Just a reminder that DJI does not recommend you to your 3rd party application when flying your DJI Drones. These 3rd party applications might affect the said DJI drone when in use. There is a possibility that these 3rd party applications will be able work. However, since it's not optimized and tested by DJI we cannot guaranteed the stability of the said 3rd party applications if it will work all the time with no issues. Thank you and have a safe flight always.

This reply is appreciated DJI Stephen.  A response in another forum suggests that I re-calibrate the collision avoidance sensors, so I'll give that a shot and report back to both forums so other Litchi users can learn from my mistakes haha.

Muttley7 Posted at 6-14 21:59
This reply is appreciated DJI Stephen.  A response in another forum suggests that I re-calibrate the collision avoidance sensors, so I'll give that a shot and report back to both forums so other Litchi users can learn from my mistakes haha.

Hi there Muttley7. Thank you for the reply and for these additional information. Have a safe and a happy flying always.

I had a similar issue with my Mavic Pro when using Litchi some time ago. The cause was a weak RC signal. As you probably know, the flightpath (waypoints) is uploaded to the aircraft, but the RC remains in control of the camera at all times. Therefore, when the RC signal becomes weak, then the camera might perform unexpected movements.
In addition, the “low noise” propellers will also cause the camera to dance under certain conditions.

MeAtHome Posted at 6-20 02:22
I had a similar issue with my Mavic Pro when using Litchi some time ago. The cause was a weak RC signal. As you probably know, the flightpath (waypoints) is uploaded to the aircraft, but the RC remains in control of the camera at all times. Therefore, when the RC signal becomes weak, then the camera might perform unexpected movements.
In addition, the “low noise” propellers will also cause the camera to dance under certain conditions.

This reply represents a major breakthrough for me, MeAtHome   Before those two flights when the camera went awry, I had recently invoked the FCC option for both my Mavics which enabled me for the first time to watch entire 6-mile round trips unfold in real-time, thanks to that vastly improved RC signal reach.

It never occurred to me that a weak RC signal could be responsible for those erratic camera movements until I read this enlightening post you shared here on the subject.  I was about to embark on a whole series of complete sensor calibrations in my quest to solve the wild camera syndrome when all I need do is switch off the controller as soon as the drone ranges out of earshot.   

Before I applied that FCC modification to both Mavics,  all footage was flawlessly recorded with a rock-steady camera view, although the video feedback typically started to break up a lot earlier during Litchi flights, so I routinely switched off my RC controller until a couple of minutes before the drone was due back at the launch point, which was generally at the 22-minute mark.

All I need do now is simply revert back to my old practice of turning the RC off as soon as the drone's hum fades off into the distance while outbound on those longer Litchi sorties.  My hat is off to you sir, for this key nugget of wisdom I have now filed with my growing folder of all things drone flying with Litchi.