Good day,
I was creating a video of a residence close to the beach for realtor friend using many flights to capture different angles and elevations. The incident was an unexpected lateral movement by Solo approximately four feet off the ground. I do not recall any problems from the prior flights of that day.
The incident occurred very quickly. When Solo appeared to go off the planned mpcc path my immediate thought was to land since Solo was close to the ground (perhaps I should have selected fly:manual). Solo was still moving laterally when the landing occurred which resulted in Solo skipping a couple of times, but remained on all fours. Solo stopped adjacent to of some ground cover and two rotors made contact for a couple of seconds before the motors stopped. There was no apparent physical damage to Solo other than some small abrasions on the rotor tips.
The flight was initiated on a vacant lane next to the beach. I established the initial cc points to fly up the middle of the lane, then fly over a dune towards the beach, and finally pan back towards the residence while over the beach and increasing altitude. After setting the points I started the mpcc from the high altitude point over the beach. Solo was flying down the lane approaching the last points when the incident occurred.
My review of the tlog in MP showed the EKF icon color on the HUD turn amber approximately the same time the satellite map view showed Solo deviating from the planned path down the lane. As best I can tell the amber color is a warning that there was a higher than expected difference between the Kalman Filter prediction and the gps system. So perhaps either a sensor(s) or the gps system may have caused the difference. The EKF alert then turned red, and the "error pos horz variance" appeared on the HUD. Interestingly, the MP satellite view showed the path deviated to the north and a greater distance than the actual path which was to the south.
The tlog also showed a drop in the parameters satellites_visible and eph (HDOP), and an increase in
"pos_horiz_variance". Ardupilot documentation says an hdop value of over 200, which I had, indicates a bad position value. It also says that a significant change in hdop and the the number of satellites can result in a gps position change. The lane is flanked by trees and bushes, but provided a clear view above.
After the incident I picked Solo up, while still powered on, to check for damage. The tlog showed some
large rolls which is likely me rotating Solo and a "bad gyro" alert appeared on the HUD. The alert went away after five seconds.
Analyzing flight data is a first for me and so I am not sure if I have a full understanding of what happened.
I would appreciate any help.
Thanks much
I was creating a video of a residence close to the beach for realtor friend using many flights to capture different angles and elevations. The incident was an unexpected lateral movement by Solo approximately four feet off the ground. I do not recall any problems from the prior flights of that day.
The incident occurred very quickly. When Solo appeared to go off the planned mpcc path my immediate thought was to land since Solo was close to the ground (perhaps I should have selected fly:manual). Solo was still moving laterally when the landing occurred which resulted in Solo skipping a couple of times, but remained on all fours. Solo stopped adjacent to of some ground cover and two rotors made contact for a couple of seconds before the motors stopped. There was no apparent physical damage to Solo other than some small abrasions on the rotor tips.
The flight was initiated on a vacant lane next to the beach. I established the initial cc points to fly up the middle of the lane, then fly over a dune towards the beach, and finally pan back towards the residence while over the beach and increasing altitude. After setting the points I started the mpcc from the high altitude point over the beach. Solo was flying down the lane approaching the last points when the incident occurred.
My review of the tlog in MP showed the EKF icon color on the HUD turn amber approximately the same time the satellite map view showed Solo deviating from the planned path down the lane. As best I can tell the amber color is a warning that there was a higher than expected difference between the Kalman Filter prediction and the gps system. So perhaps either a sensor(s) or the gps system may have caused the difference. The EKF alert then turned red, and the "error pos horz variance" appeared on the HUD. Interestingly, the MP satellite view showed the path deviated to the north and a greater distance than the actual path which was to the south.
The tlog also showed a drop in the parameters satellites_visible and eph (HDOP), and an increase in
"pos_horiz_variance". Ardupilot documentation says an hdop value of over 200, which I had, indicates a bad position value. It also says that a significant change in hdop and the the number of satellites can result in a gps position change. The lane is flanked by trees and bushes, but provided a clear view above.
After the incident I picked Solo up, while still powered on, to check for damage. The tlog showed some
large rolls which is likely me rotating Solo and a "bad gyro" alert appeared on the HUD. The alert went away after five seconds.
Analyzing flight data is a first for me and so I am not sure if I have a full understanding of what happened.
I would appreciate any help.
Thanks much
