One-dimensional vs. two-dimensional inputs on the flight controls?
 

I’m just analyzing sidestick inputs from a flight simulator study with manual flying (raw data). On Airbus aircraft you can either do one-dimensional/separate or two-dimensional/combined inputs on the two-dimensional sidestick.


Some flight trainers say that a pilot should separate the inputs on both axes, doing only one-dimensional inputs, i.e. pitch or roll inputs, but not at the same time. One possible reason for separating both axes might be a higher control accuracy when manipulation one axis only.


I’m looking for documents, reasons, literature for or against this practice. Does anyone know such documents or know some reasons for or against the separation of axes?

That would be rather limiting in a helicopter!

If you are incapable of two dimensional inputs you should stay on the ground, together with the muppet trainers who say otherwise.

PM

Well, first I would suggest those instructors go rent a Cessna and re-learn how to fly because that is the most useless statement on training I have ever heard. I agree with PM's sentiment of "muppet trainers."

Hell, the secondary effects of controls is what, Lesson 2 or 3 of a PPL...You really cannot get any more basic.

In fact, to answer your question of "some reasons...against the separation of axes" my answer would be the secondary effects of controls. Plain and simple as that. To do anything else is to forget and ignore the very basic principles of flight.

My god, what a joke our industry is becoming if this is really how pilots are taught. Any pilot who actually flies or teaches like that deserves the bottom of the barrel pay they are no doubt getting.

Reminds me of an occasion in the (Dash 7) sim, after all the serious business was complete -

LH pilot handling - "Pitch jam - Help" - Neither pilot able to move controls in pitch
"Pull pitch disconnect handle" - RH controls now operating correctly in pitch, LH still jammed
RH pilot - "I have control"

A few minutes later
RH pilot "Roll jam - pull roll disconnect handle" - LH controls now work in roll, but still jammed in pitch
LH pilot "I have control in roll only - you still have control in pitch"

Then proceeded to fly the ILS with LH pilot controlling pitch, and RH pilot controlling roll. Actually not as difficult as it seems, as you only have to correct "your" deviations.

(Apologies if this is a common simulator game on larger types, but the Dash 7 was the only type I ever flew that had pitch and roll disconnects)

(So long as it is done in the right spirit) what a fun way to wrap up a session in the bat cave ...

Never heard that one from any of my trainers. But to be honest, there might be some (very limited) credit to that theory on airbus types. After all, one does not command a control surface deflection, only a g-load command in pitch and a roll rate command in roll. The FBW system takes care of which control surfaces need to be deflected and how much to get the result the pilot wants. And no, no secondary control effects in normal law either.

But in the real world usually two dimensional inputs are needed, so the discussion is rather academic. And even the FBW system in the airbus can't paint over the fact that it is, after all, just a normal airplane, as hard as it tries to do so.

I really have no idea what’s important for flying small aircraft –and I don’t have a PPL or higher. But I assume that there are differences between what’s important for small aircraft and what’s important for the big ones. And presumably even between the lower automated Boeing types and the higher automated Airbus types.



Who of you guys is accustomed to GA aircraft and who is accustomed to Airbus airliners?

The burden of proof lies with the trainers. Like my RW brethren before have stated, you won't last long in a helicopter if you do not make combined inputs. And big aircraft are aerodynamically not that different from light aircraft. If the FBW requires single input only, it was poorly designed.

Somehow I think this is a kneejerk reaction to the AF447 continued stall; I might be wrong though. But fly the damn plane and fly the simulator the same way. There are too many scenarios where you will have to combine inputs for this to work. Before you know it you will have someone sliding off a runway in crosswind conditions because he received negative training.:ugh:

Had this one a few times on the sim recently. Quite confusing in the first few moments. I take/give the flaps and gear control to the 'pitcher' and comms and checklists to 'roller'.

I summarize:
- Airbus aircraft provide a g-load command for the pitch axis and a roll rate for the roll axis.
- The Airbus fly-by-wire flight control system compensates altitude excursions during turns.
- Thus, an Airbus pilot only has to concentrate on / control one axis during turns and can operate both axes separately – in most situations.
- From an ergonomics perspective operating only one axis is easier than two axes à the Airbus fbw approach provides reduced taskloads for pilots while manual flight.
- This is not about the incapacity of two-dimensional imputs, however, this support might erode “two-dimensional-input-skills”.

Call me oldfashioned, but this leads to a further dumbing-down of the craft of flying.

I understand that the A/P is set up to give the smoothest, most economical ride. But if the pilots move the controls and the A/P takes care of all the aerodynamic compensation, they will lose the sense of correlation between different factors and how they affect the aircraft and flight path..I don't see how this is a good idea in the long term.

As a helicopter pilot I currently fly a machine with retractable gear that has no gear speed limit. However, because some people will fly different machines in the future and some of them WILL have speed limits for u/c deployment; hence we train them to check and call the airspeed before moving the lever just to keep the awareness alive. This control system does exactly the opposite..

Apollo, I'm a bit old and thick and worse still an ex-trainer ( basic jet) but I'm struggling to reconcile that statement with landing pretty much any type, big or small, using the "crab" technique in a crosswind - where in the final moments you need (hopefully) co-ordinated inputs of roll, elevator and rudder all at the same time.

I would agree with the sentiments of +TSRA when he says

Hello Apollo,

I am a current A330 pilot. Hope the following might be of some help.

From observation in the sim, I'd agree that most experienced airbus pilots separate their longitudinal and lateral control inputs. I know that it is certainly a feature of my control strategy. Given the nature of Normal Law, the aircraft is best suited to pilots with a very 'open loop' approach. My personal view is that it is very difficult for pilots used to conventionally controlled aircraft to make the transition to Airbus FBW. There is a significant tendency to overcontrol, particularly when asymmetric. Many trainers introduce the concept of single channel control inputs into training to assist the student in going open loop i.e. 'put in the roll input on its own, and look how the pitch attitude looks after itself!'

Also, there are relatively infrequent occasions where parallel control inputs are required if the aircraft is being flown correctly. The nature of FBW means that the pilot is generally setting a target, rather than correcting a disturbance.

That said, there is no official guidance on the matter within the Airbus Flight Crew Training Manual.

Have you seen Matt Ebbatson's phd dissertation on the loss of manual flying skills in the pilots of highly automated airliners? His hypothesis might be very different to yours, but I'd suspect that the methods that he used to conduct analysis would be of some use. (simulator data, time series analysis, fournier transforms etc).

Send me a personal message if you would like to discuss further - or would like a copy of Matt's phd thesis.

Andy


Wiggy, Piltdown Man, 212 Man,

I'm afraid your comments largely indicate a lack of familiarity with Airbus FBW. The general principles of handling an aircraft are universal, but the specifics of technique and control strategy vary dramatically with type and role. I can think of several examples from previous types that I've flown. Puma and Chinook cyclic versus collective climb is an obvious one. E-2 pilots waggle their ailerons to correct vertical deviations on approach, rather than moving the yoke forward. In both cases, there are sound technical reasons why the pilots are applying this compensation to overcome aircraft handling deficiencies. The techniques are not 'wrong', just different.

+TSRA,
If you're going to post from the high moral ground, it helps to be sure of your facts first.

The Airbus in Normal Law has no secondary effects of controls.

Maybe an hour or two in an Airbus sim might help? :)

I wouldn't know about asymmetrical, but there is certainly a steep learning curve during the first few months on the line, especially during gusty conditions. And when the first year or so has passed and one does not think so much about flying the aircraft the airbus way (wait a second or two after a gust, otherwise you end up in a violent PIO) there is a chance for a relapse into overcontrolling.

Dunno about the A330 which has a more modern flight control architecture, but on the A320 there is some delay between a flightpath disturbance and the FBW realising that something happened that wasn't commanded and then doing its best to counteract that disturbance. Pilots with considerable experience on conventional aircraft tend to correct too fast for the FBW to be able to react on its own. Takes some time to learn to control only the flightpath, not disturbances.