You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
I have a PX4 MC flying indoors with external vision position (position only) system and offbord control using velocity setpoints. It was working correctly for a long time, but a few days ago we have tried a longer mission. In the middle of flight PX4 couldn't reach the desired position, it was too low. After forcing my algorithm to pass the waypoint we finished a mission and saved logs. This happened twice and both the logs have "unknown date". One log is corrupted and cannot be read by review.px4.io, but the second one is working: https://review.px4.io/plot_app?log=d5bbf9ab-effd-41ed-ba2f-d5fbba72902a
I believe the hardfaults were triggered after landing, probably during a reboot or power cycle.
The PX4 velocity estimator were indicating that the MC was flying up with 0.5 m/s even if the position was not chaining. We gave it plenty of time to climb, but it did not happen. Take a look at the 2:30 in the logs above. Z velocity is not zero but the z local position is constant. The z setpoint was limited to 0.4 m/s.
We are running a code compiled from master:
Your branch is behind 'origin/master' by 359 commits, and can be fast-forwarded.
To Reproduce
We are flying without a GPS using external vision system and offboard control. We couldn't observe it during short and low flights.
Expected behavior
We expect the velocity estimation to be correct or at least close to true values.
Describe the bug
I have a PX4 MC flying indoors with external vision position (position only) system and offbord control using velocity setpoints. It was working correctly for a long time, but a few days ago we have tried a longer mission. In the middle of flight PX4 couldn't reach the desired position, it was too low. After forcing my algorithm to pass the waypoint we finished a mission and saved logs. This happened twice and both the logs have "unknown date". One log is corrupted and cannot be read by review.px4.io, but the second one is working:
https://review.px4.io/plot_app?log=d5bbf9ab-effd-41ed-ba2f-d5fbba72902a
I believe the
hardfaults
were triggered after landing, probably during a reboot or power cycle.The PX4 velocity estimator were indicating that the MC was flying up with 0.5 m/s even if the position was not chaining. We gave it plenty of time to climb, but it did not happen. Take a look at the 2:30 in the logs above. Z velocity is not zero but the z local position is constant. The
z setpoint
was limited to0.4 m/s
.We are running a code compiled from master:
Your branch is behind 'origin/master' by 359 commits, and can be fast-forwarded.
To Reproduce
We are flying without a GPS using external vision system and offboard control. We couldn't observe it during short and low flights.
Expected behavior
We expect the velocity estimation to be correct or at least close to true values.
Log Files and Screenshots
https://review.px4.io/plot_app?log=d5bbf9ab-effd-41ed-ba2f-d5fbba72902a
Drone (please complete the following information):
The text was updated successfully, but these errors were encountered: