Unexplained event during photogrammetric mission with P4

[PeteM]

During a photogrammetric flight guided by Litchi, just few seconds after take-off (drone at the planned altitude flying towards the first waypoint of the mission) the camera (positioned towards the ground in timelapse mode) suddenly changes its angle and positions itself parallel to the ground.

There was no way to bring the camera back to the correct position during the entire mission (since in waypoint mode the drone does not respond to commands).

I had no way (yet) to repeat the mission because I was left without batteries charged... but what could have happened ?... I'm afraid it happens again ...

 

[Fly Dawg]

There was no way to bring the camera back to the correct position during the entire mission (since in waypoint mode the drone does not respond to commands).

The camera will respond to commands in waypoint mode, if your settings are not pre-loaded, which is what I believe yours were. Off the top of my head, I can't think of a reason for this occurring but what would be helpful are two things. First, share your mission plan here from the hub and second, share your CSV flight log from your device here and we can take a look to see what the issue may be.

 

[PeteM]

The camera will respond to commands in waypoint mode, if your settings are not pre-loaded, which is what I believe yours were.

I always take off manual (Litchi FPV mode) and, once the flight altitude has been reached, I do this (in order):

1) I press the C1 button to orient the camera orthogonal to the ground
2) I enter in the waypoint mode and I load the mission
3) I click on go
4) As soon as the drone starts moving, I turn on the timelapse to shoot automatically

During the last mission I did, as always, this procedure. The drone starts flying from the take-off point to the first waypoint, but after a few seconds changes the orientation of the camera in parallel to the ground.

I immediately noticed this and I started to press C1 to bring the camera back the orthogonal position, but the drone did not respond to the command. I was able to orientate the camera with C1 only when the drone completed its flight and reached the last waypoint of the mission

 

[Fly Dawg]

but the drone did not respond to the command.

Did you try the tilt wheel? Also, this is tough to answer without any data to go by. You will see the gimbal angles in the flight log and possibly a reason for the return to the parallel position, but you will need to compare those to your mission profile. This is why I suggested both. On a side note that is neither here nor there, I always load the mission before takeoff. Personal preference.

 

[PeteM]

Did you try the tilt wheel?

No. I thought about it too late...

I have the flight plan in csv format (which I load on the Litchi Hub) and I also have the flight log that Litchi automatically loads on airdata.com.

How do I get the CSV flight log from my device ? (its not the same from the airdata.com one ?)

 

[Fly Dawg]

How do I get the CSV flight log from my device ? (its not the same from the airdata.com one ?)

Yes, it should be the same as on Airdata. You can share the Airdata link here, as long as you make the CSV fully sharable, it can be downloaded. As for the mission itself, just share the link to your mission on the hub.

 

[PeteM]

In your PM you should find all the links...

 

[Meta4]

During a photogrammetric flight guided by Litchi, just few seconds after take-off (drone at the planned altitude flying towards the first waypoint of the mission) the camera (positioned towards the ground in timelapse mode) suddenly changes its angle and positions itself parallel to the ground.

There was no way to bring the camera back to the correct position during the entire mission (since in waypoint mode the drone does not respond to commands).

In normal waypoint missions the camera angle is controllable via the gimbal tilt wheel.
But why are you using Litchi for photogrammetry?
For Photogrammetry there are dedicated apps that will help you in setting out a complex flight path to fly a precise grid, fly the mission and control the photography to acquire images with the precise overlap you need.
Litchi won't do any of that.
Have a look at DroneDeploy.

 

[PeteM]

But why are you using Litchi for photogrammetry?
For Photogrammetry there are dedicated apps that will help you in setting out a complex flight path to fly a precise grid, fly the mission and control the photography to acquire images with the precise overlap you need.
Litchi won't do any of that.
Have a look at DroneDeploy.

Flights are planned, at PC, through dedicated topographic softwares. These software save the flight plan in CSV format, so I use Litchi in order to turn this file into "something" that P4 understands and executes.

In normal waypoint missions the camera angle is controllable via the gimbal tilt wheel.

Once I realized the anomaly I could have try to adjust the angle of the camera through the gimbal tilt. I did not think about it. I tried pressing, several times, the C2 button of the radio control, but without success.

Anyway...

The analysis of the log (thanks to Fly Dawg) showed that indeed a few seconds after take-off the camera tilts 90 degrees, apparently without logical reason. There is also something potentially suspicious in the middle of flight...

I'd like to find out why... I extracted the file dat from the drone and uploaded it to GDrive... I can make it available to anyone who can read it to understand what happened...

 

[Meta4]

Flights are planned, at PC, through dedicated topographic softwares. These software save the flight plan in CSV format, so I use Litchi in order to turn this file into "something" that P4 understands and executes.

Litchi has to be next to useless for real photogrammetric work.
Why would you use it when there are very good apps that do photogrammetry missions perfectly?

 

[PeteM]

I begin to understand what could have happened...

The button C1 of the radio control performs the action that is pre-set within the menu of Litchi and NOT, in general, the orientation of the camera.

Since I have always pressed the C1 button in FPV mode (immediately after the manual take-off and before opening the waypoint menu and loading the flight mission) I have always believed that the C1 button was used to orient the camera.

So when, during flight, I realized that the camera was no longer oriented towards the ground, I started to press C1... but... in the WAYPOINT mode the button C1 has no longer the "Reset Gimbal" function (as in FPV mode) BUT the "Waypoint at Aircraft" function...


Reset Gimbal
For the Mavic/Phantom 3-4/Spark: toggles the gimbal between horizon and looking down. For the Inspire 1-2: resets the gimbal yaw.

Waypoint at Aircraft (Waypoint)
In Waypoint mode, adds a waypoint at the aircraft location. The newly added waypoint will have its altitude, heading and gimbal pitch angle automatically set to the current aircraft altitude, heading and gimbal pitch angle.


And this explains why the camera was "locked" in its wrong position during all the flight...

It remains to understand why the camera has changed direction a few seconds after the start of the auto mission... could I inadvertently press some rear buttons ?... like C1 ?... or like C2 (POI at device) ?

 

[PeteM]

Litchi has to be next to useless for real photogrammetric work.
Why would you use it when there are very good apps that do photogrammetry missions perfectly?

I use Mission Planner.

Litchi is just the solution to "drive" the drone with the output generated by Mission Planner...

 

[athearnsd45]

Also note that you can exit Waypoint mode by flicking the flight mode toggle on the left side of the controller to a different mode momentarily, then back. That way you won’t have to burn battery flying a mission you’re not happy with. More importantly, it allows you to exit a mission that puts the drone or others in danger. This switch is on the left of my P4A controller; I assume it’s in the same place on a P4 controller.

 

[PeteM]

Today I repeated the mission and I solved the mystery of why the camera has changed angles.

In the new version of Litchi they have implemented the command to orient the gimbal displayed on the screen in such a transparent way that it can not be seen and when I did tap the screen to zoom the map I passed my finger over this command.

The rest (because I could not move the camera anymore) was explained in the previous post.

Problem solved.

 

[Fly Dawg]

Problem solved.

Excellent!

 

[PeteM]

Also note that you can exit Waypoint mode by flicking the flight mode toggle on the left side of the controller to a different mode momentarily, then back.

I think that from now on I will never forget it.

So far I had always flown without problems and when I saw this anomaly (and the camera that did not respond to the radio control) I panicked and I was forced to wait for the end of the flight (burning the last remaining battery and relaunching the flight to another day...)