I tried an DO SET ROI auto mission , planning my waypoints more or less in a circle around a point of interest :
The take off point (home) is in the circle, so the drone should first get to the first waypoint by travelling more or less along the circle radius line, then once it reaches the first waypoint, it should go to the second waypoint, etc.
However what happens is the drone indeed flies along the radius line just after take off but then goes on way too far before doing a kind of spline curve to get back to waypoint 2:
Then I tried another auto mission and again it did these splines curves between waypoints:
I am flying with Pixhawk, firmware 3.1.2 which normally does not support spline curves.
So why does the drone seem to fly spline curves anyway ?? or could there be another cause to these strange paths between waypoints ?
Thx for any help on this behavior
Replies
Before the problems started, this is what mine used to do:
I've since become aware of waiting a good 5-10 minutes for GPS lock, and now wonder if changes in solar radiation can affect GPS. Something changed causing my Y6 to act very goofy for a few days even with good hdop. Just sitting on the ground it would wander around several meters whereas before it stayed right there. I've been reluctant to fly missions because of it. Nothing in the logs suggested a mechanical error or glitch of some sort (except once when I mucked up a parameter setting). Do post your log though, it may reveal the issue.
It looks always good (except maybe at the very beginning slightly above 2)
I am going to post the onboard video on youtube for you to watch. You will notice int his video another strange phenomenon on waypoint 18 : the X8 just pauses for a long moment (5 to 10 seconds) before resuming the mission.
In both cases it nails the later GPS coordinates. So is one possibility that it is locking onto more sats as it flies and getting more precise. One way to determine this is to fly the course twice, one right after the other, landing and disabling, but not turning off power to the ship. Then see if the two flight paths overlap. If they do, then you know it's in the software and not external.
Very good point. What is the HDOP value before take off? You should be less than 2.0. Also, how is the magnetometer? Has it had a MOT tune done?
Have you changed the extended tuning settings for nav?