Plane: proposed changes for 4.6.0-beta5 #29531
Conversation
if the GCS commands a guided re-position with an AGL frame and quadplane is in Q_GUIDED_MODE=1 then this prevents us adding the home altitude to the target terrain altitude
ensure that period[] array is not updated until we push
ensures we can recover from inverted flight quickly
when we start the VTOL motor stabilisation with an attitude beyond normal attitude limits we set a flag to use fixed wing attitude control (slaving VTOL rate control to fixed wing rate control) until we have recovered this avoids an issue with the shaping in the VTOL attitude controller and also fixes an issue with the VTOL controller bringing the nose down, which can cause a lot of height loss in quadplanes
we use input_rate_bf_roll_pitch_yaw_no_shaping to avoid input shaping when slaving the VTOL rate controllers to the fixed wing rate control this allows for much faster recovery in Q-assist
when we are close to a 180 degree error, go with the last direction if we are in a 40 degree angle wedge of 180 degree error
we need to fly somewhere with a different terrain height than home or we cannot test terrain target
|
Hasn't made it to the call yet but I would suggest #29494 too. Cork push should be fine, will only affect DShot I think? Copter has the extra complication of the fast rate stuff of course, that will need testing. VTOL recovery one make me abit nervous, removing that rate liming could result in vehicles crashing in normal operations if there FW time constant results in a higher gain than the VTOL angle P. A halfway step might be to fix the reset to measured altitude to be reset to FW target, then they both start out in the same place and the accel limit will have much less impact. Guided changes look fine. |
I don't understand what you mean, sorry. I'm guessing you mean attitude not altitude, but when are you thinking we do this? you mean extra calls to attitude_control->reset_target_and_rate() ? |
This rolls up some key PRs that are being considered for beta5.
I have test flown the 3rd change myself. The other two have only been SITL or bench tested so far.
The fast attitude recovery PR needs more consideration as it will impact VTOL control during a forward transition. It fixes an issue where VTOL rate control does not match fixed wing control due to the input shaping in the VTOL control. Before considering this for beta5 we need some flight tests to look at any possible impacts
I will create a custom build tag for this PR to help users do testing