Hyperlapse Waypoint wrong altitude

[speedpilot]

Good afternoon,

I want to make hyperlapse video (mini 3 pro) every few weeks of a house construction. Preferably with the waypoint to make a really cool effect.

Because of a YouTube video I found out that with saved way points the altitude (at least the first waypoint) is not the same. I’ve jus tried this and happend to me too.

I saved a few waypoint positions and every time I flew to a random spot and hit record. It flew back to the same position at 20 meters altitude to start the hyper lapse; Perfect. Then I landed, turned of the drone and back on. Up in the air and started to same saved waypoints. It flew to the same position but now at 14 meters…

And guesses? Thank you.

Stefan

 

[akdrone]

Manually adjust it is all you can do Although I admit I have not tried that so I don’t know if that is actually a viable solution

 

[Dave Maine]

Remember the waypoint altitude is not the height above ground, or absolute altitude. It is above the height where the drone was powered up and launched. Launch from a different point and you can end up at a different height above ground. Also note that the barometric altimeters used are not very accurate or repeatable.

 

[akdrone]

I think the point being made is that after you save a set of waypoints and repeat it, the drone does not return to the same starting height as before. I have heard this reported before. It's seemingly a known thing even if you repeat a waypoint set immediately after with (probably) no change in barometric pressure.

 

[speedpilot]

I think the point being made is that after you save a set of waypoints and repeat it, the drone does not return to the same starting height as before. I have heard this reported before. It's seemingly a known thing even if you repeat a waypoint set immediately after with (probably) no change in barometric pressure.

Yes this is correct it seems to be a software problem.

 

[speedpilot]

I've tried it again today.
First flight, setting up few waypoints. First one at 23 meters. Then started to record the hyperlapse and it went back up to the exact position I planned at 23 meters.
Landed the aircraft, powered down, powered up and took off again (same position as first flight, also waited for the homepoint to be updated before I took off). Loaded the saved waypoints and pushed record. It went back up to the correct position, but now at 25 meters. And the 3rd time I did this it was 26 meters...

This was all within 30 min.

 

[Harryscopic]

I think that's about as good a precision as you can get from this drone with satellite positioning. Also, when recording the waypoint, the altitude display is blocked by the hyperlapse pop up window so you can't see how high it is, at least on a Samsung S10e phone.

 

[speedpilot]

The horizontal position is spot on every time. Just the altitude is off, if it would just use the altitude that is showed on screen it would be perfect every time. (To clarify the drone is physically also higher when it shows the 25 meters, I used the top of a building as reference) So it is super accurate if it used this altitude…

Also today I tried it again, ended up at 28 meters. And physically it was much higher then yesterday. It’s a shame.

 

[HarleyFlyer]

All my testing has shown the same - it doesn't like anything lower than about 15m as a starting point as it tries to go into the ground - and it seems to be +-10m out for each waypoint. I am hoping that it will work ok for my 9 waypoint - across sunrise/noon/sunset and those across 4 seasons - will let you know in a years time...

 

[mobilehomer]

Altitude is set by barometric pressure. That pressure constantly changes. Normal atmospheric changes are far greater than that perceived by the sensor. Also, GPS altitude variance is on the order of 10-15 meters.

 

[speedpilot]

Altitude is set by barometric pressure. That pressure constantly changes. Normal atmospheric changes are far greater than that perceived by the sensor. Also, GPS altitude variance is on the order of 10-15 meters.

This doesn't explain the wrong altitude. Yes baro pressure changes, but within a few minutes not much has changed on a normal sunny day. DJI drones use baro for altitude and also the on screen altitude (at least thats what I read everywhere)

The problem is, the altitude indication is correct all the time. The building Ive used as reference is 23 meters. Every time I fly to 23 meters (the indication on screen) it is pretty much spot on with the top of the building. But somehow the waypoint starts higher (different every time), in reference to the building and the indicating that on the screen too. So it must be using a different altitude. Almost like it maybe uses GPS altitude for the waypoint, which isn't shown anywhere.

 

[HarleyFlyer]

This doesn't explain the wrong altitude. Yes baro pressure changes, but within a few minutes not much has changed on a normal sunny day. DJI drones use baro for altitude and also the on screen altitude (at least thats what I read everywhere)

The problem is, the altitude indication is correct all the time. The building Ive used as reference is 23 meters. Every time I fly to 23 meters (the indication on screen) it is pretty much spot on with the top of the building. But somehow the waypoint starts higher (different every time), in reference to the building and the indicating that on the screen too. So it must be using a different altitude. Almost like it maybe uses GPS altitude for the waypoint, which isn't shown anywhere.

I believe the GPS altitudes are being recorded on the exif data for photos. So when I look at my hyperlapse folder and examine the exif data there I can clearly see an altitude above sea level shown.

 

[speedpilot]

I believe the GPS altitudes are being recorded on the exif data for photos. So when I look at my hyperlapse folder and examine the exif data there I can clearly see an altitude above sea level shown.

Interesting, I've send the question to DJI too.

 

[jeremytipp]

Good afternoon,

I want to make hyperlapse video (mini 3 pro) every few weeks of a house construction. Preferably with the waypoint to make a really cool effect.

Because of a YouTube video I found out that with saved way points the altitude (at least the first waypoint) is not the same. I’ve jus tried this and happend to me too.

I saved a few waypoint positions and every time I flew to a random spot and hit record. It flew back to the same position at 20 meters altitude to start the hyper lapse; Perfect. Then I landed, turned of the drone and back on. Up in the air and started to same saved waypoints. It flew to the same position but now at 14 meters…

And guesses? Thank you.

Stefan

Did you ever find a good solution for this? I just began construction on my home and I'm hoping to do the same thing and I'm wondering if you ever found a good solution?

 

[akdrone]

GPS is notoriously inaccurate with regard to altitude. The problem is a simple one of triangulation of the GPS signal and there may not be anything that DJI can do about that. This is just guesswork on my part but I have over the years noticed a significant differences in the GPS altitude and this may be another instance.

 

[Dave Maine]

For the repeated flights, did you start from exactly the same launch point. The altitude the drone flies to is really the height above the original launch point, not altitude above ground or sea level.

 

[Droning on and on...]

Altitude is set by barometric pressure. That pressure constantly changes. Normal atmospheric changes are far greater than that perceived by the sensor.

Not what i learned in Atmospheric Science in college.

Large pressure changes take place over many hours, unless there is a front moving through, and even that depends on how tight and organized the front is, as well as how fast it's moving. Barometric altimeter are pretty accurate and stable over the span of a drone flight.

Wind correlates strongly with pressure changes. If it's calm and still, pressure will be very stable.
 
However, accuracy isn't in question here. The altitude returned in telemetry is different than the recorded waypoint altitude. It should read the same as that recorded, no matter how inaccurate it might be.

Something else is wrong other than sensor accuracy. A waypoint was recorded at 23m, and the Mini3P is reporting it is flying at a different altitude than the recorded waypoint.