He to all,
Just two line of presentation, i’m a pilot and i own an ultralight aircraft . I’m also a programmer and i love the integration from hardware and software as done here. So thanks to all the arduino bulder.
Now i have this question: has anyone tried all or part of the ardupilot on a real aircraft?
I have an idea on my mind and for my airplane: to install an autopilot without changing drastically my airplane controls. I have trim on two axes and here come the second question: can an airplane be controlled by using trims? (Roll and pitch)
I tried inflight and seems to work (manually) what you think about?
Hoping 
for any opinion
Massimo
I am also a pilot in fixed wing aircraft and helicopters. I am not familiar with trim control in ultralight planes, but in certified aircraft trim is only used to relieve control pressure in stable flight so the pilot does not tire from holding constant control pressure in one direction, just like trim used in RC aircraft. Trim is never intended to control the aircraft for many reasons, the most prominent being that not enough control input can be delivered to correct for things like wind gust or turbulence.
In my opinion, what you are suggesting would be extremely dangerous. Say you were able to hook up servos to your trim and APM and actually got it to work. You would be the equivalent of a test pilot on an RC plane. Having crashed enough RC aircraft in my “test flights” the odds are not in your favor.
The APM is a great piece of technology, but I don’t think I would trust my life to it on a real aircraft. I’m pretty sure that’s not what it was intended for.
You are right, but moving the trim (i have a switch on the stick) i’m able to add a force to the control surface so the surface itself move in the opposite direction and when the attitude is reached relieve it by moving the trim again.
I dont want to crash of coarse and i’m shure that some kind of pressure on the control surface and of coarse on the stick cannot become so dangerous.
The idea is only to stabilize the aircraft and to mantain the heading.
PS: trims on ultralight works in the same manner as on a cerifyed aircraft. I have one on the aileron and one on the elevator both electric so the only needs is to intercept the wires on behind the panel.
Massimo
I think the trimming would be too slow, and could create oscillations as the plane won’t be able to keep up with the APMs demands.
I think it could be done safely if you have some sort of clutch that will release the APM control of the trim tabs, and you have the strength to overpower the trim tabs at Vne when they are trimmed fully in each direction.
Testing this would be the most dangerous part.
Yes, the problem is that trims are slow, they take few seconds to reach the position and more they are not step motor so i must deal with time.
About the first problem i think that the mouvement back ( return to the original position) must be anticipated by a prederined number of degree.
I mean if the attitude is wrong of + 5 degree i start moving the trim until the attitude error stop then calculating the angular speed and the time to reach it i think to be able to start the trim before the error is = 0 so when the error is null the trim must be in the original position ( before the error)
Dont warry, for the test i already have a switch for trims rear and left side and i’ll made tests on the ground befor in flight 
Massimo