Thanks for the pointers! @losawing
I will try to increase rate P gains even more and observe the results, then play with D.
I did increase Q_ANGLE_MAX to 80 degs, and hovering there both forward and backward was no problem at all (while in q_stabilize)!
After reading about the attitude controls I suspect AC_PosControl is the next main objective while in q_hover mode?
Along with the parameters shown in the last image here: https://ardupilot.org/dev/docs/code-overview-copter-poscontrol-and-navigation.html
I’ve already set Q_M_THST_HOVER to at least one correct significant digit. But I suspect having one more would be preferred. I did set it after yesterdays crashes, which would explain the first switch to seem like falling out of the sky. The logs indicate a 50% throttle upon switching, that increased, but too slowly.
Will also set Q_M_SPIN_MIN today.
Looking at my last logs and just by eyeballing the elevons on the ground when in FWBA the controls look rather agressive. Given my oversized elevons is it correct to assume that the plane parameters, such as PTCH2SRV, etc, should be reduced quite a lot from the standard firmware settings?
This seems counter intuitive, I know flying wings with single engine don’t have yaw control and work fine with arduplane, but when given the option why not? Just curious!
This I need to check out, thanks!
Thanks again!
@kd0aij@losawing
FLTMODE list does cover most of the modes I see, but I also see 10 as ARMED, is this my mistake or is it possible that the matlab logs contain more/other info as well?
I haven’t tried to make a “board” fly in FW modes. I guess it should work if you have the CG in right place, but stall characteristics might be quite interesting
Thank you very much for your contribution. My vector two-axis aircraft (bicopter) has been able to fly normally. Now I want to add ultrasonic height determination and optical flow positioning, but I can’t use ultrasonic and optical flow. There is no flow enable parameter for optical flow in the parameter table.(FLOW_ENABLE)
Thank you very much for your contribution. My vector two-axis aircraft (bicopter) has been able to fly normally. Now I want to add ultrasonic height determination and optical flow positioning, but I can’t use ultrasonic and optical flow. There is no flow enable parameter for optical flow in the parameter table.(FLOW_ENABLE)
Here a new version of VTOL controlling. https://youtu.be/2UEVRb5xpLc
The 45° Mode allows to fly forward against wind where the plane is not pressed down due to the airflow on top of the wing because the whole plane has to tilt.
I red about a branch where this was tested for Tilt-Rotors in ArduPlane but not continued. https://www.flexinnovations.com/product-p/fpm3870.htm
Ups, I seemed to be late. In the vid it looks perfect in SITL
What about a little bit backwards to brake?
My Skywalker needs 100 m until it hovers on spot.
I hadn’t thought about reverse throttle for tiltrotors, but that should be easy to add.
The interesting part would be integrating it with the existing reverse capability…
Do you mean to reverse in FBW Mode?
During the Transition from Hover to FBW the trust changes from vertical to horizontal while accelerating.
The same would mean for the back Transition in Throttle reverse against the fly direction to tilt down for Hover while decelerating.
But then the tilts are in the opposite direction for the next forward Transition.
Difficult to explain
Yes, I was thinking about backward tilt in FW modes.
Applying some backward tilt during the back transition would be independent of that, so it would be simpler to specify.
Using vectored Yaw for my Skywalker, the Tilts are able to tilt backwards max. 25°.
But may be, then also the Backtransition becomes difficulties to mangage the Yaw. The Forward Transition produce a Yaw of about 30°always to the right. The Tilts don’t move synchronously.
And I did not finish the Transitiontuning because of a PC crash after a Win 7 Update. The Transition takes a long time to accelerate, one time with increasing the Throttle manually only. Really bad compared with the same Wing as Dual Mot Tailsitter.
Arduplane : Tailsitter 4.00
EKF3
I have some problem about AUV. Actually the altitude of my AUV is reduced but the altitude from QTUN is increased which is not related to altitude from BAR2. Can be seen from the graph shown. How should I fix this problem?
Log : Uploading: 2020-01-11 14-54-47.bin…
I had a similar situation. This is due to the sealing of the compartment with the flight controller. when speeding up, pressure or vacuum was created in the body of the aircraft
How did you solved it?
Baro’s altitude is sensitive while the aircraft is transiting to FW from pressure changing. So if i change EK3_ALT_SOURCE from baro to GPS, aircraft will transit better. But when vtol landing, it will land in altitude inaccuratcy because GPS antenna is not point to the cloud.
I understand right?
hi guys
iam new to pixhawk and want to try build a non-vectored tailsitter
is there a parameter to be my reference?
and how about servo and esc wiring ?
@devres A good source for example parameters is the SITL_models repo:
Find a model there that is similar to yours, and start with its parameters.
Probably the best example for a non-vectored duo-motor tailsitter is Tridge’s here:
