RTK position and accuracy from RTK enabled UAVs

WebODM
#1

Hi all,

I’m just wondering if there’s somewhere I can set the positional accuracy of the georeferenced photos I’ve captured with the M300 RTK we have. I know in another package, it reads the location from the photo, but also the quality of the position. Being RTK, the fixed position of the photo is restricted to a smaller positional uncertainty. If there’s a flag or setting which adds these values to the adjustment, it would be greatly appreciated if you could let me know.

Thanks.

1 Like
#2

Hi.

It is my understanding that the gps-accuracy tag is intended for that.
https://docs.opendronemap.org/arguments/gps-accuracy/#gps-accuracy

Still, a clarification on the following quote would be welcome: “If your images are tagged with high precision GPS information (RTK), this value will be automatically set accordingly.”

1 Like
#3

Thanks very much for the reply. I’m going to try setting this tag with the expected accuracy of the RTK system. The DOP resolution is included in the EXIF data as another package seems to be reading it in, but we had a bit of a bowling effect on some data that we processed. The error suggested this was not being read correctly. I’ll give an update after manually setting it.

1 Like
#4

Hi Craig

I’m also flying the m300+p1 and still fine tuning the accuracy. Right now i am seeing best results with gps-accuracy 0.4. Smaller than that brings out a scale problem.

Testing the sfm triangulation algorith is next on my list.

Also i havent had much success with smart oblique yet. I think the backwards pitch and resulting orientation require a unique camera profile.

Happy to compare notes…

2 Likes
#5

Hi!
Thanks for the post regarding the accuracy. We’re only a couple of days into it so haven’t had too much of a play to have a decent amount of notes. The orthoimage produced appears to be quite good which is one of the applications we intended to use it for. We’ll keep playing with the accuracy and see if we can some better results. Haven’t even attempted the smart oblique function yet.
Thanks.

#6

Let me know what you learn! I did try to test the triangulation algorithm and got an error that the gimbal angles weren’t in the EXIF data. They clearly are there so perhaps some work needs to be done to connect the dots. Maybe we need a separate M300+P1 discussion thread!