I am having 3 problems which I cannot get rid of. Maybe anyone can help.
1. I'm quite frustrated with the Ctrl-F geotagging method. I was never able to get all photos correctly tagged and have no idea why and what I'm doing wrong, although following exactly the procedure as described in
http://diydrones.com/profiles/blogs/geotagging-images-with-mission-planner
Usually I do a few test photos before take off, which I guess shouldn't confuse the offset estimation. There's always several photos not tagged and a good amount is tagged totally wrong. Strange. I definitely need a way to geotag my photos, as 1000-2000 photos are no fun to align in photogrammetry without any camera position.
2. I observe height losses of at least 10 meters up to 20 meters in turns with my plane which looks similar to the X8. I would expect the plane to compensate the expectable height loss when turning with the elevators much earlier than it actually does. Is this a malfunction, an unknown behaviour or bad parameters tuning?
3. A really irritating observation I made when the mission is finished and the plane returns home, as it rapidly loses height and would definitely crash if it weren't stopped by manual control. I couldn't find a reason for this, though I assume to simply have forgotten taking care of some parameter somewhere.
Can anyone help in these problems?
Many thanks.
Oliver
Replies
1. You can apparently enable the "CAM" log line on the APM board (better probability it's logged, since telemetry is not 100% reliable). This only works if you're using pin A9 for triggering. It will include position and angles.
You can use alternative methods if you have a couple of reasonable positions or GCP's. Usually the photo positions are only valid to give a rough idea where the photo was taken, because in post-processing they can try to match photo features of other photos in a certain sphere around it. This reduces processing time.
The actual geo-referencing is then done through GCP's, which is *always* preferable. GPS can have consistent errors of 2 meters or so due to atmosphere propagation. The GCP's need to be realy precise with specialized equipment, but even using a phone is better than relying on a GPS from a moving vehicle because the time between trigger and actual photo shutter release is unknown. That error only disappears when you fly in all main directions around the target area.
2. Not sure. Not apm plane user.
3. Do you have relatively high wing loading? then reducing speed for landing preparation can cause a severe drop.