Recovery from unusual attitudes is required during primary training and must be demonstrated during the checkride. In the case of an incapacitated or disoriented pilot, will activating a single-axis autopilot in ROL mode while in an unusual attitude safely recover straight-and-level flight?

  • $\begingroup$ A lot of autopilots have a "straight and level" button which can marginally recover from an unusual attitude. You have to be aware though that the autopilot isn't a pilot, and its "recovery" methods may make the situation worse, or be unable to recover from all attitudes, especially if altitude is not permitting. $\endgroup$ – Ron Beyer Jan 5 '16 at 16:55
  • $\begingroup$ Also I think a disoriented pilot may be able to push this button and turn the autopilot on, however if in the case of an incapacitated pilot, its doubtful a non pilot rated passenger could activate the autopilot and select that mode without some training. Some autopilots also have an 180° turn button that holds altitude for inadvertent flight into IMC. $\endgroup$ – Ron Beyer Jan 5 '16 at 16:59
  • 3
    $\begingroup$ Maybe I'm reading your question too literally, but a single-axis autopilot couldn't recover to straight and level flight because that would require altitude hold in addition to roll control. Or did I just misunderstand what you mean? $\endgroup$ – Pondlife Jan 5 '16 at 20:52

You would need a three-axis autopilot. The nature of an unusual attitude is that it needs corrections around all three axes.


Generally, no. There's always an exception out there someplace, but I've never seen an autopilot that is advertised as being able to recover from an unusual attitude situation. Depending on exactly what the attitude is, the mode of the autopilot, and its specific capabilities & mechanical limitations, you MIGHT be able to get out of an unusual attitude that way, but I wouldn't count on it. And it's certainly never taught that way anywhere that I've been... you're taught to recognize-confirm-recover by flying out of the unusual attitude yourself, rather than relying on the automation to do it for you. (In fact, if you get into an unusual attitude, chances are decent that either mismanagement of or a failure in the automation contributed to getting you there!)

Given the wide range of possibly unusual attitudes & aggravating circumstances (i.e. engine out, rudder out-of-trim, shifting or shearing winds, structural damage, failed pitot/static instruments, failed gyro, etc etc), I'd be very surprised if any manufacturer would ever claim that their autopilot could be relied upon for this task.

  • 2
    $\begingroup$ The Cirrus autopilot (an Avidyne DFC90) is advertised as doing exactly that. But it isn't single-axis, of course. $\endgroup$ – Pondlife Jan 5 '16 at 20:49
  • $\begingroup$ Very cool -- hadn't seen that capability before. I wonder what the certification regime was in order to get that capability proven and approved? I wonder also what sort of envelope that capability is good for (inverted + nose high, for instance). $\endgroup$ – Ralph J Jan 5 '16 at 21:04
  • $\begingroup$ The documentation says it works from an "arbitrary" starting position, but it's only been demonstrated up to +/-60 bank. Although at that point, I think the general idea in a Cirrus is to pop the chute. $\endgroup$ – Pondlife Jan 5 '16 at 21:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.