Coordinates inaccurate from takeoff

[slup]

You are correct that the points are displayed on the map properly, both home point and landing point and the course of the flight. But the coordinates describing those points are inaccurate. They are both out west of Taos Pueblo, not on the west slopes of the Sangre de Cristos. Plot those coordinates on the USFS or geological service maps and see where you are. Mobilehomer found the correct coordinates in the KML file.

You must have made some mistake when you transferred the coordinates to the geological maps you speak of ...

You say that this is a correct plot of the flight you did in the right place..? This is taken directly from what you shared from Airdata in the first post (the Airdata .KML file downloaded & shown in Google Earth).


Below is the .CSV file from your flight (the Airdata .CSV downloaded & shown in Excel). Have marked the last known coordinate on the last row there...



And here I've dropped a red pin with the last coordinate from the .CSV file above in Google Earth ... & the pin perfectly line up with where the flight path ends.

 

[oshamtn]

You must have made some mistake when you transferred the coordinates to the geological maps you speak of ...

You say that this is a correct plot of the flight you did in the right place..? This is taken directly from what you shared from Airdata in the first post (the Airdata .KML file downloaded & shown in Google Earth).

View attachment 155756
Below is the .CSV file from your flight (the Airdata .CSV downloaded & shown in Excel). Have marked the last known coordinate on the last row there...

View attachment 155757

And here I've dropped a red pin with the last coordinate from the .CSV file above in Google Earth ... & the pin perfectly line up with where the flight path ends.

View attachment 155758

You must have made some mistake when you transferred the coordinates to the geological maps you speak of ...

You say that this is a correct plot of the flight you did in the right place..? This is taken directly from what you shared from Airdata in the first post (the Airdata .KML file downloaded & shown in Google Earth).

View attachment 155756
Below is the .CSV file from your flight (the Airdata .CSV downloaded & shown in Excel). Have marked the last known coordinate on the last row there...

View attachment 155757

And here I've dropped a red pin with the last coordinate from the .CSV file above in Google Earth ... & the pin perfectly line up with where the flight path ends.

View attachment 155758

I will say this once again. The flight path as it appears on this and the several other screens that folks have displayed is accurate. It's accurate enough that I blew it up, traced the roads (thanks to Google satellite view, though Google mislabeled the applicable forest road: 478 should be 441. The terrain is recognizable, and other roads are labeled correctly. (A mistake of this nature is understandable, as 441 was part of the tertiary fileline for the huge Hermits Peak/Calf Canyon fire that burned 534 square miles this spring in Northern New Mexico and has been altered). I drove up there the day after the drone was lost, brought my Garmin GPS unit and was dumbfounded when I looked at the coordinates from the CSV file. They are not anywhere close to this mountain top. They are on the Taos plain. When the KML file produced coordinates, they matched the map and the route. I went back up there today and hiked down to the trail below 441 (441-4, which is closed to vehicles). I began with 36.17.32 and started walking east from 105.25.10. I found the drone at 36.17.509N, -105.25.042W, on the trail. The drone appears to be in good shape despite the hard freezes at night. I called DJI support and asked their advice. They told me to do a test flight and if I had problems, make arrangements to send it in. I did a test flight and had no issues, but it was only a few hundred yards. Auto takeoff, auto landing, camera, all good -- so far. I asked them about the CSV file. They said that happens occasionally, and told me it is usually associated with the need for a firmware update. I checked, and my firmware is still up-to-date. I'll send an email tomorrow and include more detail and hope to gain some useful information. If so, I will post it.

 

[Meta4]

I began with 36.17.32 and started walking east from 105.25.10. I found the drone at 36.17.509N, -105.25.042W, on the trail.

It looks like you are confusing the location format.
The numbers you were given from the recorded data are degrees and decimal degrees, but the numbers you are showing on your GPS and quoting are degrees, minutes & seconds.
This is why you were confused.

I did a test flight and had no issues, but it was only a few hundred yards. Auto takeoff, auto landing, camera, all good -- so far.

It should be ok, since it autolanded, not crashed.

I asked them about the CSV file. They said that happens occasionally, and told me it is usually associated with the need for a firmware update.

DJI support people have no idea.
If they don't have a script to answer your question, they suggest updating firmware.
Ignore them.
The explanation for your confusion is described above.

 

[Elthree]

36˚ 17.509' N, -105˚ 25.042' W
is the same as
36.291827˚N, 105.417367˚W
People don't put multiple decimal points in a number for a reason: it's ambiguous!

Note that the last known location mentioned above was 36.291838˚N, 105.417327˚W.
I would call that pretty close! By my reckoning that's about 4m away!

Also, the negative longitude implies West. If we were being literal a negative west coordinate would be a positive east coordinate. In the middle of China!

Generally I set all my equipment to use the same coordinate format, but it's still important to recognise when something is presenting a different format.

 

[Meta4]

I drove up there the day after the drone was lost, brought my Garmin GPS unit and was dumbfounded when I looked at the coordinates from the CSV file. They are not anywhere close to this mountain top. They are on the Taos plain.

Same location expressed in decimal degrees and degrees minutes and seconds

 

[oshamtn]

It looks like you are confusing the location format.
The numbers you were given from the recorded data are degrees and decimal degrees, but the numbers you are showing on your GPS and quoting are degrees, minutes & seconds.
This is why you were confused.


It should be ok, since it autolanded, not crashed.

DJI support people have no idea.
If they don't have a script to answer your question, they suggest updating firmware.
Ignore them.
The explanation for your confusion is described above.

Thank you, Meta4. That explains the discrepancy. And I appreciate learning something I didn't know.

 

[tvdwouw]

I think every drone flyer should read this post. GREAT.