Delaying pitch-up for heavier planes may be beneficial for obstacle clearance. (getting to highest angle airspeed quicker).
1 Like
Hi @menschel,
Sure, thanks. Added to the 4.6.0 issues list in the āPlane specific issuesā section.
thanks for the report! It isnāt a watchdog reset, it is āmain loop stuckā, which is related but not quite the same thing. A main loop stuck means that the main loop didnāt run for 500ms. If the main loop doesnāt run for 2000ms it does a watchdog reset, but in your case it recovered.
Can you see if you can find the full tlog for this issue? Iād also like to know which parameter you changed (that should be in the tlog).
If you were running with LOG_DISARMED then the bin log would be good too.
Is the issue reproducible on your setup?
I see what you mean, @menschel . Thanks for the report!
Iāll try to reproduce this in SITL, and make a regression test.
Then fixing it should be easy enough.
Ground speed makes no sense for an aircraft equipped with an airspeed sensor. You can have quite a bit of airspeed in headwind.
Weāve just released 4.6.0-beta6 and the announcement including release notes is here.
There will be a -beta7 to resolve an issue with the LD19 proximity sensor but besides that, this -beta6 is very close to the final version that will become the 4.6.0 stable release.
All testing and feedback is greatly appreciated!
Today I updated 4 planes (3 VTOLs, 1 fixed wing) from beta 5 to beta 6 and had a nice flying day without any problems (Matek F405wing, H743 wlite, Q-Modes, Auto, Cruise, FBWA)
Rolf
1 Like
We released plane-4.6.0-beta6 last week. My apologies for the late announcement! Many thanks to those that have already been testing it.
The changes vs -beta5 are in the Release Notes and copied below:
- Board specific changes
- ARK FPV, LumenierLuxF765, MicroAir743, StellarF4v2, X-MAV-AP-H743v2 OSD Type defaults fixed
- BETAFPV-F405-I2C UART6 fix
- BrotherHobby F405v3 and H743 support
- FlywooF405HD VTX power pin fixed
- NarinFC-H7 support
- TBS Lucid H7 Wing support
- TBS Lucid H7 support for both I2C ports
- Driver bug fixes and enhancements
- DroneCAN semaphore bug fixed which affected MPPT battery and Serial-over-CAN
- GPS logs altitude above ellipsoid instead of undulation
- GSOF GPS protects against port misconfiguration
- UBlox GPS configuration sped up
- Lua script battery monitor failures that could cause hardfaults fixed
- Nova GPS undulation fix
- Proximity backends increased to 5
- SCHA63T IMU temperature reporting bug fixed
- Plane specific fixes and enhancements
- Circle mode terrain alt handling fixed
- Fence re-enable after fence breach fixed
- Prevent rotation pitch calculations from running post-rotation
- Takeoff leveloff timeout check added
- Terrain guided target intermediate alt handling fixed
-
Copter SCurve navigation fix to avoid excessive vertical speed when passing through waypoints
-
Rover circle mode and QuickTune support smaller vehicles
-
Bug Fixes and minor enhancements
- AHRS initial orientation fixed when autopilot mounted in non-default orientation
- AIRSPEED mavlink message health flags fixed
- CAMERA_FOV_STATUS message always sent (once requested), attitude is earth-frame and FOV can be decimal numbers
- DDS / ROS2 fix to provide clock_gettime only on ChibiOS boards
- Lightware I2C and HC SR04 fixed timestamp for last reading (should only affect AP_Periph devices)
- āNo ap_message for mavlink idā warning removed
- Power brick2 valid flag fixed on some boards
- Pre-arm check that gyro read rate is at least the main loop rate
There will still likely be one more release (e.g. -beta7) to resolve an issue with the LD19 proximity sensor but besides that, this beta6 release is likely very very close to what will be in the official release so all testing is greatly appreciated!
@menschel I really appreciate all your bug reports, but can you make sure you include a log with each report like this? I know it may seem obvious for you what you are talking about, but it is less obvious for me. Ideally you would reproduce the issue in SITL and include a link to a bin log from SITL showing the problem. That makes it extremely easy for me to then reproduce and work on the issue. It doesnāt surprise me that you are finding issues with the large changes in terrain handling in 4.6.0 as it was a very large code change, but we canāt just revert as it was fixing serious bugs.
If you donāt have a place to upload bin logs to then let me know and Iāll create a folder for you on my google drive.
The graphs from MAVExplorer are useful too (thanks!) but ideally I need the bin log associated with each issue, so I can extract from that log the exact parameters, mission and GCS commands (they are all logged in the bin log, allowing me to reproduce easily)
Thanks!
1 Like
You should probably know that I currently have over 1.2 million unread emails. Please tell me what email address you are sending to for me, and what exact email address you are sending from. That at least gives my email search a fighting chance to find your emails 
3 Likes
Hi
VTOL transitions:
There seems to be a bug, or a wrong parameter description, in Plane 4.6.0 Beta 5 and 6 (these are the versions we have tested).
Documentation says Q_ASSIST_SPEED = -1 disables the feature. BUT it also disables back transitions from FBWA to Q_STABILIZE. Transitions from Q_STABILIZE to FBWA works and uses AIRSPEED_MIN as documentation says. Previous behaviour was that AIRSPEED_MIN was also used for back transitions.
Tested and verified with beta 5 & 6 on two different VTOLs.
What camera and vtx is that? Beautiful analog video
Hi @pkvaloy
Just to understand, so we donāt get confused:
If the transition back from FBWA to Q_Stabilize does not work, did you land the VTOL in fixed-wing mode? Or what exactly do you mean?
Iām asking because I canāt reproduce this with SITL (Mission Planner) . Could you link a logfile ?
I have flown several VTOLs (and fixed wing) with 4.6.0 beta 5 and 6, but VTOLs always with Q_ASSIST enabled. If I donāt misunderstand you, I will fly a real VTOL tomorrow with Q_ASSIST off at -1.
Rolf
@pkvaloy
The transition FBWA->QSTABILIZE works fine with Q_ASSIST_SPEED -1 Plane 4.6.0 Beta 6
Perhaps other conditions must be met for the error you described to occur. A log file or at least a parameter file might clarify this.
Rolf
3 Likes
@menschel Iāve reproduced and fixed the NAV_LAND with terrain alt bug.
I expect to include this in the final of 4.6.0, or next beta
Thanks very much for reporting this!
Hi, thanks for looking into this. Our RC-pilot swithes to QSTABILIZE when the transition doesnāt occur automatic.
Thanks a lot for looking into this
Log file from tests with a Fighter 3.0, Arduplane 4.6.0 Beta 6:
bin log
Parameters
Test were done at about high (100m) altitude. Parameters were changed during the tests to reproduce the behaviour. Some back transitions work (where Q_ASSIST_SPEED different from 0 or -1). When Q_ASSIST_SPEED is set at 14 m/s back transitions work fine.
Best regards - PĆ„l KvalĆøy
@pkvaloy
Q_Assist works, but the activation message doesnāt appear, which is irritating. The quad motors are already running before your pilot switches to Q_STABILIZE, as shown in the log file.
