Forum

Author Topic: Problems with kml import of the detailed flight plan created in Metashape!  (Read 475 times)

Vermesser

  • Newbie
  • *
  • Posts: 6
    • View Profile
Hello everybody!

I need help to successfully import the flight plan created in Metashape (current version) into a flight planning app!

In the past few days I tried unsuccessfully to find a solution on the Internet. I also tried several APPs. Many promise kml import, but often you can only import a simple ring or something similar. We have come the furthest with the IOS Litchi APP for DJI. After importing, the drone flew the path and the photos were released, but unfortunately the drone was not aligned according to the flight plan and the gimbal was not controlled either. So the result is completely useless. What are we doing wrong? Such an optimized flight plan with all the configuration options is actually a wonderful thing! Does anyone have any idea what we're doing wrong? Does anyone even have their own experiences that could help us? We are currently testing this as part of a study project at our company.

I would be very grateful for your support!

Have a nice week
Patrick

simiyutin

  • Newbie
  • *
  • Posts: 13
    • View Profile
Re: Problems with kml import of the detailed flight plan created in Metashape!
« Reply #1 on: February 24, 2021, 10:02:31 PM »
Hey!


You may try using DJI Pilot app instead if Litchi. It officially supports only a small fraction of drones but quietly works with a larger subset (e.g. regular mavic). You need to export plan in DJI Pilot KML format in this case.

Also you can try using Android version of Litchi - it seemed to work fine in the past

Vermesser

  • Newbie
  • *
  • Posts: 6
    • View Profile
Re: Problems with kml import of the detailed flight plan created in Metashape!
« Reply #2 on: February 25, 2021, 04:46:58 AM »
Hi!

Thanks for your very informative hints!
I will share the new information with our student. We should definitely try these variants.

It sounds like you have already gained experience? If so, does the data exchange and the later flight work smoothly for you? How Do the results look like? I would be interested to know whether the automatic flight planning is convincing in practice? Let's see if we can achieve results.

Thank you very much for your support!

simiyutin

  • Newbie
  • *
  • Posts: 13
    • View Profile
If your drone supports RTK then positioning and orientations are very precise, it is possible to set capture distance as low as 1 meter from the surface and safety distance as low as 0.5 (in case of well-reconstructed obstacles or exhaustively specified by manual tools), but you still have to carefully follow the drone in case of GPS shadowing in concave regions.

When only GPS is available positioning is prone to errors up to several meters and it is recommended to set safety and capture distances at 15 meters or more.

For some examples you can check out model built from 3500 flight planned photos and screencapture of another model with flight plan overlay



Vermesser

  • Newbie
  • *
  • Posts: 6
    • View Profile
Hello simiyutin!

Thank you for your additional information! I'll have a look at it in a moment. For this student project we only had a variant without RTK. It is and was interesting for me to see what is possible with a hobby UAV. Otherwise we have an octacopter with a system camera at our company. This is not so handy, but the results are very good. In europe, the drones will be reclassified soon, although no exact details have yet been published, so there are still no classified UAV on the market. A DJI P4P RTK is actually very interesting for such things. But it has been on the market for a relatively long time, so the future should be questionable. In the future, we would need UAVs that are classified as safe as possible. In the company jobs also take place in an inhabited area. I find the topic of automatic flight planning for building detection really very interesting. Let's see what the future brings! A new classified UAV with RTK and camera for photogrammetry would be cool, of course cheap;).

Thank you for your kind and very informative support!