Has there ever been any aircraft of any type that could glide safely with no control input simply based on the way the frame of the aircraft was designed?

I was trying to think of requirements for building an aircraft that would be accessible to the masses and the first thing I thought of was safe landing in the event of engine failure or other catastrophic mechanical breakdown.

It seems to me that it would be possible using modern composites to create some type of aircraft whose default flying configuration is a safe, slow descent. NOT including parachute equipped such as cirrus, has any such craft been designed/flown?

  • 16
    $\begingroup$ Descending safely is easy, it's landing safely that's the real challenge :-) No amount of aerodynamics will help you if you glide straight into a brick wall. $\endgroup$
    – Pondlife
    Commented Sep 29, 2014 at 20:45
  • 9
    $\begingroup$ A well designed paper airplane exhibits those characteristics. $\endgroup$
    – abelenky
    Commented Sep 29, 2014 at 21:49
  • 4
    $\begingroup$ @abelenky: A well designed paper airplane exhibits a Reynolds number higher than any existing manned aircraft. $\endgroup$
    – dotancohen
    Commented Sep 30, 2014 at 10:33
  • 3
    $\begingroup$ You might also want to look into AutoLanding AutoPilot. If you are wanting to design a system to be "fool-proof", I'd say a huge component would be computer controlled/assisted systems. $\endgroup$ Commented Sep 30, 2014 at 16:48
  • 4
    $\begingroup$ @FreeMan haha! Indeed. That reminds me of a quote I saw once that said, "Programming is a race between programmers to produce bigger, better, more idiot-proof systems and the universe to produce bigger and better idiots. So far, the universe is winning." I think this could be generalized equally well to all of engineering. $\endgroup$
    – reirab
    Commented Jul 22, 2015 at 19:20

6 Answers 6


Broadly speaking most (if not all) light GA aircraft can "glide safely with no control input" - aircraft are generally designed to have positive dynamic stability, such that they will return to a stable equilibrium condition (e.g. "level cruise flight") in the face of most modest upsets. Once configured for cruise flight they can maintain it with little input from the pilot (and if equipped with even a basic autopilot "little input" can often be reduced to "no input" for extended periods of time).
Whether or not the engine is producing power is largely irrelevant here, save for the fact that if the engine isn't producing power you will eventually be descending.

Positive stability alone will not make aviation "accessible to the masses" however, nor will autopilots: As with driving a car or riding a motorcycle there are certain "aeronautical decision-making" skills which a pilot must possess in order to safely fly (and land) an aircraft when everything goes right. If you introduce problems (like engine or instrument failures) the need for a real live pilot becomes even more critical: technology can not yet replace the critical decision making capabilities of a trained human mind.

Particularly in regard to your question about a default flight mode of a controlled descent, even a simple autopilot can already do what you describe: Planes will continue to fly the last autopilot-commanded heading or route until they run out of fuel, at which point they'll start descending (while still attempting to fly the programmed profile).
What they cannot do is select a suitable landing site, update their flight profile, and fly themselves to a safe landing. For that to happen without human intervention requires a huge amount of luck.

In a particularly famous example of this kind of luck a Piper Comanche "landed" itself in a field after running out of fuel, the pilot being unconscious at the time due to carbon monoxide poisoning. The pilot survived and in fact literally walked away from the accident scene, and the aircraft in question is, as best I can determine, still flying.

  • 2
    $\begingroup$ I vaguely recall there was a plane that landed intact in a corn field after the pilot baled out or ejected. Must try to find that. $\endgroup$ Commented Sep 29, 2014 at 23:23
  • 11
    $\begingroup$ en.wikipedia.org/wiki/Cornfield_Bomber The "Cornfield Bomber" was a Convair F-106 Delta Dart, operated by the 71st Fighter-Interceptor Squadron of the United States Air Force, that made an unpiloted landing in a farmer's field in Montana, suffering only minor damage, after the pilot had ejected from the aircraft. The aircraft, recovered and repaired, was returned to service, and is currently on display at the National Museum of the United States Air Force. $\endgroup$ Commented Sep 29, 2014 at 23:45
  • 1
    $\begingroup$ "technology can not yet replace the critical decision making capabilities of a trained human mind." So... no Google self-flying planes in the near future? $\endgroup$
    – Michael
    Commented Sep 30, 2014 at 4:39
  • 1
    $\begingroup$ @Michael: I would say self-flying planes are very much in the near future. Perhaps marginally more likely is self-flying helicopters, as choppers are more manoeuvrable but harder to pilot safely. $\endgroup$
    – Phil H
    Commented Sep 30, 2014 at 13:09
  • 4
    $\begingroup$ @Michael We have self-flying (and even self-landing) planes now - the question is what happens when something goes wrong (e.g. the engine stops turning)? Flying from point A to point B and landing on a guaranteed-clear runway is pretty complicated, but finding a spot big enough to put your aircraft when you've got no thrust and all you can do is descend is substantially harder :) $\endgroup$
    – voretaq7
    Commented Sep 30, 2014 at 16:05

Not really a plane, but Autogyro are really safe even when motor shutdowns. http://en.wikipedia.org/wiki/Autogyro

For having tried one, when motor stops, it just descends slowly (maybe 1m/s max), and you just have to make little corrections before landing, to convert vertical speed to horizontal speed. It's really like an "air bike"

  • 4
    $\begingroup$ Welcome to Aviation:SE! Re "Not really a plane": The OP used aircraft multiple times, not airplanes. I think you've nailed a really good example in your answer. $\endgroup$
    – CGCampbell
    Commented Sep 30, 2014 at 14:04
  • 1
    $\begingroup$ The autogyro is a great example of an "inherently safe" design: it'll come down gently (even without an autopilot) in most circumstances. You'd still want a pilot to make sure it lands in a convenient spot though: Winding up in the middle of a lake wouldn't be good if there's a conveniently-empty parking lot nearby... $\endgroup$
    – voretaq7
    Commented Sep 30, 2014 at 18:29
  • 1
    $\begingroup$ Indeed, during an early test of a Pitcairn autogiro (as they spelled it), one of the rotor blades came loose and departed the aircraft. The test pilots, feeling the strong vibration and assuming the craft was about to disintegrate, bailed out. However, it flew in circles for several minutes and then landed itself in the corn field they were flying over with only minor damage to the landing gear. $\endgroup$
    – Skyler
    Commented Jun 29, 2017 at 18:20
  • $\begingroup$ True for the classic autogiros of very low disk loading. A present-day gyro, in vertical autorotation, will hit the ground at about 1500 fps. Survivable, but barely, and quite a crash... $\endgroup$
    – xxavier
    Commented Mar 26, 2020 at 13:56

You might want to look at the Antonov An-2. The leading edge slats are spring-loaded. If your airspeed falls below about 40 mph, they deploy. With them deployed and minimal/no power, your speed drops to about 25 mph and your "sink rate" is so low that the aircraft can perform a controlled descent and landing without damaging the gear (depending on the surface you land on, naturally). The plane has no defined "stall speed," meaning that you can pass out, pull the controls full aft, run out of fuel and make a relatively uncontrolled descent, yet you might actually survive.

This seems to be the closest to what you're asking for.


You are in some regards introducing a few questions here

I was trying to think of requirements for building an aircraft that would be accessible to the masses

Bringing aviation to the masses is only in part related to the aircraft its self. For what its worth the Piper Cherokee 140 can be had for less than most new cars and has what has been called "Very Benigin" handling some might even say making it to safe. The plane like most GA planes glides very easily and practicing engine outs is routine to get your license. Which brings up the more important point. The current US regulations set fourth by the FAA makes it much tougher to fly a plane (although sport pilot has helped this) than it does to drive a car. To be honest it does not take that much to legally drive a car here in the US. Learning to fly an airplane is a significantly larger task, it takes an investment (anywhere from 7K-15K depending on local flight school costs) and more time than learning to drive a car. This is really what keeps the masses from flying.

For what its worth new planes are not cheap not only because they are expensive to build but because FAA type certs are not easy to come buy.

and the first thing I thought of was safe landing in the event of engine failure

Engine failure is, (assuming there is no fire involved) a recoverable situation as long as there is an open field or even a local parkway to put the craft down. Small planes (and most planes for that matter) glide fairly well and can be maneuvered under glide with ease.

or other catastrophic mechanical breakdown.

The problem with this is that most of the clean handling characteristics of small GA planes are predicated on the fact that the plane is as it should be. There are a million and one things that can break but may will cause a situation that will alter the handling or flight characteristics of a plane like asymmetric flap deployment or snapped control cables/disabled flight controls or even critical air frame failures like the 2008 incident involving a Pilatus PC-6.

could glide safely with no control input simply based on the way the frame of the aircraft was designed

As has been mentioned most GA planes (at least the slower trainers) are made to return to a safe attitude to some extent. It should be noted that this often requires some kind of minimum altitude and many accidents happen at low altitude when the plane simply does not have time to recover. Some of this may also be effected by things like trim which if set to its limits has an effect on the planes handling.


Aircraft for the unskilled pilot have a long history. Provided the plane is basically stable, there are several conditions for a safe dead-stick landing:

  • Wing loading must be low. This will keep the rate of descent to a safe level, so the pilot does not need to flare up and slow the descent for landing.

  • Aerodynamic efficiency must not be too high, so that the plane does not "float" in ground effect for a long distance.

  • For true hands-off, you have to avoid phugoid oscillation. Flown in this condition, many planes will slowly begin to nod slowly up and down, the oscillations eventually growing to disastrous levels.

First off the block was the tailless swept-wing biplane developed by British pioneer J W Dunne. It was also capable of automatically recovering from a stall - assuming you could somehow get it into one. In 1910 it became the world's first certified stable airplane, an event witnessed by none other than an astonished Orville Wright. The Burgess-Dunne was built under license in the US and from time to time other Americans such as Smith and Waterman based their "safety aeroplanes" on it. Its main undoing is that, being tailless, it has a small CG range, which requires some experience to live with.

There are plenty of more conventional types around. Perhaps the great "everyman" classic was the de Havilland Moth biplane of the golden age.


Has there ever been any aircraft of any type that could glide safely with no control input simply based on the way the frame of the aircraft was designed?

Absolutely. While many aircraft have inadequate roll stability to truly descend safely with no pilot input, especially if some turbulence is present, some aircraft do have an ample amount of roll stability as well as pitch stability. An example would be a Rogallo-style hang glider with ample sweep to the leading edges. Placing the point of connection between the pilot's body and the aircraft structure well below the aircraft CG helps to lower the effective the CG of the glider-pilot system which further increases both pitch and roll stability via the "pendulum effect", yet hang gliders have also been witnessed to fly remarkably stably when accidentally launched off a hill with no pilot attached.

Powered hang-glider-like aircraft ("trikes") can be, and have been, built on the same principles. This miniature radio-controlled model of a Rogallo-wing "trike" is extremely stable and easily capable of descending for a prolonged period, even in turbulent air, with no pilot input. Note that in this case the "trike" unit with motor and batteries is rigidly fixed in place (the position only changes when the servos move), so the "pendulum effect" is even further enhanced-- the CG of the aircraft is far below the wing.

(The "pendulum effect" is based on the fact that the drag vector, and the aerodynamic sideforce vector generated by a sideslip, both tend to exert stabilizing torques when they act above the CG of the aircraft or the aircraft-pilot system. The "pendulum effect" has also been caused the "keel effect", though this is somewhat misleading as the stabilizing action of buoyancy on a boat with a weighted keel is independent of sideslip. Some people dislike the term "pendulum effect" as well because it implies that the aerodynamic center of the aircraft is acting as some sort of a fixed pivot point, about which the weight vector or the G-load vector exerts a torque-- this is not really an accurate depiction of what is going on.)

Note that when a pilot hangs by a single flexible strap and exerts no force with his arm muscles, his body weight acts as if it is located at the point where the strap connects to the aircraft. Some early hang gliders had this connection point located well below the "keel tube" to enhance pitch and roll stability; this practice has now been discontinued because it the resulting short "hang strap" increases the muscle force that a pilot must exert to shift his weight a given distance to the side. On the other hand, in paragliders the pilot hangs by multiple suspension lines which act essentially like rigid struts due to the triangular geometry involved; in this case the pilot's body weight no longer effectively acts as if it were located at the point where the lines connect to the wing, but rather at its actual location-- which places the CG of the whole system far below the wing and creates a powerful "pendulum effect" which leads to strong roll stability, despite the anhedral geometry of the actual wing.

Paraglider pilots have often flown in clouds using only minimal instrumentation, such as a magnetic compass. Clearly this is only possible in an aircraft with strong intrinsic pitch and roll stability. On February 4 2007 paraglider pilot Ewa Wisnerska, flying without oxygen, was unintentionally lifted into a thunderstorm and survived an accidental climb in a thunderstorm to 32,000 feet above sea level followed by a descent back to earth. She was unconscious for over an hour, with her body encased in ice. When she regained consciousness her aircraft was in a stable descent.

Of course, there is an entire discipline of model-airplane-flying called "free flight". These models rise up, and then return to earth, usually landing safely, without any control input of any kind. It is instructive to examine their configuration. They almost invariably have dihedral and rather small vertical fins, and usually have a high-wing configuration.

Related ASE questions and answers:

(Q) What is the Keel Effect?

(Q) Does "pendulum effect" apply to hang gliders or any aircraft?

(A) Does "pendulum effect" apply to hang gliders or any aircraft?

(Q) Why are high-wing aircraft more stable?

(Q) How does the "pendulum effect" affect biplanes??

(Q) Does the dihedral effect happen during coordinated flight?

  • $\begingroup$ Very brave to mention "pendulum effect" in spite of much passion of some folks reaction to it. May I suggest "dihedral effect" works better, as even an anhedralled hang glider will have its center of drag above the CG, and also sweep, resulting in a net "dihedral effect". +1 answer otherwise. $\endgroup$ Commented Mar 26, 2020 at 16:13
  • $\begingroup$ Several mythical effects have been ascribed to the poor pendulum. At least this answer names it after a valid effect. $\endgroup$ Commented Mar 27, 2020 at 15:05
  • $\begingroup$ @RobertDiGiovanni -- consider the paraglider case specifically. I prefer to avoid "dihedral effect" because in a sideways flow the upwind wing actually has a lower angle angle-of-attack than the downwind wing, due to the anhedral shape of the actual wing, and likewise with some hang gliders with visible anhedral. I prefer to keep the effect of sideforce (and drag force) acting above the CG as a separate thing, which might legitimately said to contribute to a larger combined effect called something like "effective dihedral". $\endgroup$ Commented Mar 27, 2020 at 15:44
  • $\begingroup$ @RobertDiGiovanni -- so in short I guess I would prefer to say "effective dihedral" includes the effects of various things such as actual dihedral (a geometry that induces a difference in angle-of-attack during wings or other pairs of matched surfaces during a sideslip), and sweep, and "pendulum effect" or "keel effect" (i.e. the effect of sideforce and drag force acting above the CG). I think we've talked about this before; I better stop now or the comments will get moved. $\endgroup$ Commented Mar 27, 2020 at 15:46
  • $\begingroup$ @quiet flyer you are correct in distinguishing dihedral/anhedral and side force drag torque so ... +1 for your work. $\endgroup$ Commented Mar 27, 2020 at 20:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .