yaw torque issue in large vectored tailsitter

flexible2g said:
Please provide link to real world model, none vectored, none ruddered model, to show that what you wish to do in the worlds best model simulator, which does a very good job of emulating real world aircraft.
Prove me wrong. Take thrust vectoring, giros, and rudder off a model, you DO NOT HAVE YAW CONTROL!
Flaparons,do not controll yaw. Under the best of condishions, whitch is not realistick in the real world. The only thing flaparons can do is twist the air frame untill the wing is vertical, and then you still have no yaw controll for landing.
NO FUN!
Click to expand...

I believe that the model that they are talking about has hardware in the loop connected to realflight with the Flightaxis Link. It is basically a computer controlled RC aircraft. Realflight 8 now allows this type of connection to actual flight hardware to control the sim.

I have a quad that I built with a Pixhawk which runs a version of ardupilot for quads. It is capable of fully autonomous flight without any control inputs. I myself am very green when it comes to this technology myself but it's very interesting to learn about these new capabilities in what can be done with actual RC hardware.
 
Last edited:
flexible2g said:
I am all in for what they are trying to accomplish. I can find nothing, like this hovering without vectoring. Elevons can not control yaw.
Click to expand...

https://youtu.be/o_95BNzNWnM

In that video is that aircraft I posted about holding a hover with only elevons. It seems to work. It may be using variable thrust as well but I am not sure. The op may have more information on how exactly it works.

It might be worth noting that a human pilot is not in direct control of that aircraft. It's holding its position by GPS lock and is computer controlled. It's probably able to do this because it can react much faster than a human pilot. I harken this to the fly by wire setups in modern fighter jets and commercial aircraft.
 
Last edited:
flexible2g said:
I am all in for what they are trying to accomplish. I can find nothing, like this hovering without vectoring. Elevons can not control yaw.
Click to expand...
Then you just haven't tried very hard in your searches. These are standard vehicle types in the world of autonomous vehicles.
I did suggest you lookup the wingtra. Here is the direct link: https://wingtra.com/
There are many thousands of others.
Elevons control yaw fine, both in the real world and in realflight
 
csgill75 said:
https://youtu.be/o_95BNzNWnM

In that video is that aircraft I posted about holding a hover with only elevons. It seems to work. It may be using variable thrust as well but I am not sure. The op may have more information on how exactly it works.
Click to expand...
The control system is very simple:
  • roll is from differential thrust on the two motors
  • pitch is from elevon pitch
  • yaw is from elevon (what would be roll in fixed wing flight)
  • climb is from average thrust
The key for the airframe is the very large elevons to give sufficient control. Also a willingness to pick up the aircraft off the ground when it falls over on landing :)
Detailed discussions of the ArduPilot implementation of the control scheme is here:
https://discuss.ardupilot.org/t/dual-motor-tailsitters/15302
plenty of flight videos linked in that discussion
 
flexible2g said:
You left out one very important thing differential thrust. Add differential thrust to you're model and it will fly great.
Click to expand...
yep, though it was more a matter of terminology. For a tailsitter in hover we (the ArduPilot community) call yaw the axis you get with elevon (ie. body frame roll). We call the axis you get with differential thrust roll. This is because with ArduPilot in control it makes the users inputs map to those of a multicopter when hovering, and changes to fixed wing control axes when you switch to a forward flight mode. In RealFlight8 the XPlusOne does the same thing in Hover mode.
So I thought you were saying that roll about body frame doesn't work, which it obviously does :)
I'll add in some differential thrust on rudder in the CatManual model. For the ArduPilot controlled model we use two completely separate channels for the two motors, but that would be rather inconvenient for a manual pilot, so best to mix it with rudder stick.
Sorry for the misunderstanding.
Cheers, Tridge
 
You must log in or register to reply here.
Back
Top