[HN Gopher] If a pilot ejects, what is the autopilot programmed ...
       ___________________________________________________________________
        
       If a pilot ejects, what is the autopilot programmed to do? (2018)
        
       Author : avestura
       Score  : 68 points
       Date   : 2025-10-30 21:27 UTC (1 days ago)
        
 (HTM) web link (aviation.stackexchange.com)
 (TXT) w3m dump (aviation.stackexchange.com)
        
       | ortusdux wrote:
       | I recently learned about the Green Ramp disaster, where the crew
       | ejected from an F-16 under full afterburner, and the jet
       | continued on to collide with several parked airplanes, resulting
       | in 24 fatalities.
       | 
       | "As of 2025, this incident has the largest number of ground
       | fatalities for an accidental crash of an aircraft on U.S. soil.
       | It was also the worst peacetime loss of life suffered by the
       | division since the end of World War II."
       | 
       | https://en.wikipedia.org/wiki/Green_Ramp_disaster
        
       | tekla wrote:
       | TLDR: It does nothing and it should never do anything
        
       | darkhorn wrote:
       | I think the most proper thing for the jet should be to destroy
       | itself. In a war enviornment I would not like my enemy to gain
       | intel about my military jets.
        
         | yjftsjthsd-h wrote:
         | Not my wheelhouse, but doesn't that involve packing the
         | aircraft with explosives, and wouldn't that involve risk of
         | blowing up if someone else shoots you? Or is there some better
         | way to self-destruct?
        
           | SoftTalker wrote:
           | Impacting the ground usually does it.
        
           | DonHopkins wrote:
           | Jets are usually full of jet fuel that can blow up if someone
           | else shoots you, or they run into the ground.
        
           | klysm wrote:
           | Assuming there is an autopilot present, nose diving at max
           | afterburner straight down is probably going to yield a
           | similar result
        
           | estimator7292 wrote:
           | One could calculate the amount of energy in a given amount of
           | explosive and smashing into the ground at high velocity. I'm
           | too lazy to do so, but I can tell you they're on the same
           | scale.
           | 
           | A heavy object moving fast has a _shocking_ amount of energy.
           | When such an object impacts the ground, all that energy has
           | to go somewhere.
        
           | cosmicgadget wrote:
           | There are probably a number of clever and failsafe ways to
           | divert jet fuel somewhere that would destroy the plane on
           | command.
        
           | miki123211 wrote:
           | You could self-destruct just the software / computers by
           | encrypting everything with a few keys that are easy to
           | destroy.
           | 
           | This is a standard technique for quick and secure data wipes,
           | particularly on e.g. modern SSD drives where a traditional
           | wipe might be incomplete due to wear leveling algorithms.
        
         | jojobas wrote:
         | Would you go near a plane that's an electronic signal away from
         | blowing itself up?
         | 
         | Even if mechanical, warplanes get combat damage, and having a
         | system like that could make a difference between survivable and
         | sure death.
        
           | dylan604 wrote:
           | Maybe bring some strong jamming equipment to prevent the
           | electronic signal from being received?
        
             | jojobas wrote:
             | The signal can originate on the plane itself due to a
             | software glitch, stray currents in the grounding skin or
             | whatever. If anything, jammers tend to interfere with the
             | carrier's electronics itself.
        
         | crazygringo wrote:
         | I think its high-speed collision with the ground or ocean
         | generally takes care of destroying it. Especially with no pilot
         | attempting to keep it level and slow it down and minimize
         | damage.
        
         | mikkupikku wrote:
         | You could erase computer memory, but blowing up the plane won't
         | stop them from gaining intel about the materials, mechanism,
         | and whatnot that make up the plane. An explosion won't vaporize
         | the airplane, just break it into smallish pieces. Those can be
         | collected and analyzed to reconstruct most any detail about the
         | construction of the airplane. They even do this with missiles
         | and bombs. Even shell fragments. When artillery shells packed
         | full of high explosives go off, their intricate fuse mechanisms
         | are left remarkably intact.
        
       | the__alchemist wrote:
       | If it's a controlled ejection scenario, you try to fly to a
       | specific location, airspeed, heading, and altitude, then pull. It
       | will be in your local-area in-flight guide. The intent is, the
       | plane ends up somewhere away from civilization. This if, of
       | course, only suitable for scenarios where you have this luxury.
        
       | hex4def6 wrote:
       | It seems like the sensible thing to do would be to fry / erase
       | any IFF and encryption related stuff, but otherwise continue as
       | before.
       | 
       | E.g, if it's already been programmed to fly straight and level,
       | continue to do that. If it's deactivated, stay deactivated.
       | 
       | Just seems like a whole 'nother set of characteristics to test
       | otherwise, as well as adding extra unpredictability. The aircraft
       | is probably damaged / on fire, so its flight characteristics are
       | already going to be extremely different to normal. The best thing
       | in the moment may be to let the aircraft lawn-dart in a field,
       | rather than attempt to get straight and level, and in the process
       | potentially fly over inhabited area or towards a friendly set of
       | aircraft / buildings / vehicles.
        
         | pdonis wrote:
         | If the autopilot is engaged, the pilot won't be ejecting,
         | because the aircraft will be in some kind of controlled flight.
         | Autopilots will be disengaging and lighting up a big red light
         | in the cockpit well before the aircraft gets to the point where
         | the pilot would consider ejecting. Remember that ejecting is an
         | absolute last resort, since the pilot is quite likely to be
         | injured and runs a significant risk of being killed in the
         | process of ejecting.
        
           | cluckindan wrote:
           | About one in 20 ejections results in death, usually due to
           | low altitude, or being hit/crushed by the seat.
           | 
           | Compare to 20 in 20 jet airplane crashes resulting in death
           | and suddenly pulling that lever might seem a worthwhile risk
           | to take
        
             | pdonis wrote:
             | _> About one in 20 ejections results in death_
             | 
             | But more than that result in injury. The possible injuries
             | are pretty severe.
             | 
             |  _> Compare to 20 in 20 jet airplane crashes resulting in
             | death_
             | 
             | Crashes if the plane is totally uncontrollable, probably
             | yes.
             | 
             | But there's a lot of gray area in between "totally
             | uncontrollable" and "controllable enough that an autopilot
             | can fly the plane". There are plenty of cases where a pilot
             | was able to make a controlled enough crash that they walked
             | away from it, even though the plane itself was totalled.
             | 
             | And once we get to the point of "controllable enough that
             | an autopilot can fly the plane", the pilot would have no
             | reason to eject--because the plane is controllable enough
             | that the autopilot can fly it. Which means whatever
             | problems exist can't be very severe--or the autopilot would
             | be disengaging, because it needs things to be working
             | pretty well to fly the plane at all. That was the point of
             | my response in the GP to this post.
        
           | LorenPechtel wrote:
           | Disagree--it can be entirely sane to eject from a fully
           | functional aircraft.
           | 
           | 1) Something is going catastrophically wrong with the plane.
           | It's flying now, but soon that fire will burn through the
           | hydraulics, safer to punch while it's still in controlled
           | flight.
           | 
           | 2) There is not enough fuel to put it on a runway. Once
           | again, punch while it's still stable. We've even seen a
           | civilian do that--ferry flight with one of those planes with
           | an integrated parachute. He had a fancy rig with extra fuel
           | in the cabin, it wouldn't feed. Without it he wasn't going to
           | make Hawaii.
           | 
           | 3) We have seen a Navy pilot correctly punch from a fully
           | functional aircraft. He was on final when somebody launched
           | an SM-2 at him. Low, slow, defenses off--no chance, he
           | punched. He was pointed right at the carrier at the time,
           | having an autopilot do something else would be a good idea.
           | (There would always be a chance that the missile was
           | destructed in time.)
           | 
           | And with semi-functional planes:
           | 
           | 4) We have seen an Israeli pilot bring their bird back with
           | one wing. He had to land very hot but there was a long enough
           | runway, he was able to do it. But what if it's even worse?
           | There can be enough damage that your minimum airspeed is
           | above any runway you can reach or above what your wheels can
           | handle. Or maybe the Navy would prefer the pilot to eject
           | rather than risking a major mishap on the deck.
           | 
           | 5) What if the problem is elsewhere? We have seen a pilot
           | punch from an apparently-functional F-35. The problem was
           | actually in his controls. (Yes, he has been found wrong--the
           | first two investigations cleared him (if the plane is not
           | responding properly to pilot inputs below 6,000' AGL, eject),
           | they finally found a panel that would declare him wrong.
           | Doesn't make him wrong.)
        
       | SoftTalker wrote:
       | A pilot would only eject if the aircraft was uncontrollable with
       | no reasonable hope for recovery. Unlikely the autopilot can do
       | anything deliberate at that point.
        
         | quotemstr wrote:
         | Yes, but the autopilot should have some kind of contingency
         | programming in case the pilot is mistaken about the aircraft
         | being unflyable.
        
           | appreciatorBus wrote:
           | If this was possible it would just be part of regular flight
           | control laws and would be used to avoid becoming
           | uncontrollable in the first place.
        
             | quotemstr wrote:
             | Huh? If I'm the human pilot, I can pull the ejection lever
             | for multiple reasons, including my just being an idiot. The
             | plane, after I eject, should do _something_ reasonable.
             | Maybe it
             | 
             | * starts broadcasting a mayday?
             | 
             | * crashes into the nearest large body of water?
             | 
             | * attempts to fly itself back to base (we have the
             | technology)?
             | 
             | I mean, it has to do _something_ and flying straight and
             | level until it runs out of fuel is unlikely to be the
             | optimal value of  "something"
             | 
             | Why would it be controversial to say "Look, guys, we should
             | decide what the plane does after the pilot ejects. Maybe
             | the best policy _is_ just flying same course and speed
             | until fuel exhaustion, but we should choose this policy,
             | not default into it without consideration. "
        
               | SoftTalker wrote:
               | Ejecting for for no reason would end the pilot's flying
               | career. Ejecting for _any_ reason will result in an
               | investigation, at minimum. Pilots are expected to fly the
               | airplane until the last extremity.
               | 
               | So while yes it's possible, it's unlikely, and the return
               | on investment of making the plane able to do something
               | like "return to base" in that circumstance would be a
               | large negative number.
        
               | throwup238 wrote:
               | Even ejecting with good reason is enough to end a fighter
               | pilot's career. The rates for significant back injury are
               | between 1 in 3 and 1 in 2 depending on the design.
        
               | mikkupikku wrote:
               | My understanding is they're automatically retired from
               | flight after the second (or maybe third?) ejection, not
               | automatically after the first.
        
               | appreciatorBus wrote:
               | It feels like a good chunk of the comments here imply
               | that ejection is a pretty common activity, hence the
               | interest in what happens next, both to the pilot and the
               | jet. I don't have any ready statistics, but if ejection
               | was a commonplace activity then the world's air forces
               | would be depleted of planes in very short order. It
               | wouldn't shock me if there's a policy about pilots who
               | ejected more than x times, but I would be shocked if the
               | policy was ever actually enforced, simply due to the
               | rarity of it even being an issue.
        
               | throwup238 wrote:
               | The Martin Baker homepage shows 7802 lives saved using
               | just their brand of ejection seat so it's not exceedingly
               | rare.
               | 
               | I'd estimate less than 5% of fighter jet pilots eject
               | during their career.
        
               | WheatMillington wrote:
               | Do you realise we're dealing with humans? Humans who make
               | decisions based on a multitude of factors, or sometimes
               | none at all?
        
               | SoftTalker wrote:
               | Yes, but flying aircraft with ejection seats is demanding
               | work and few humans are capable or qualified to do it.
               | Most people can barely manage to drive cars safely.
        
               | ratelimitsteve wrote:
               | remember that part of optimization is the amount of
               | resources spent developing a solution for a problem that
               | just doesn't come up that often. in the microcosm of a
               | single ejection there's probably a better way to handle
               | it than to just let the plane continue on its course. in
               | the macrocosm, there's probably better problems to deal
               | with than the one that results from the relatively rare
               | situation in the military and unheard of in the civilian
               | sphere. it's also worth noting that ejector seats are
               | explosive-assisted and any plane that's been ejected from
               | is rendered structurally unreliable, and usually is so
               | close to crashing that nothing can be done to save it
               | even if saving it is viable. So most of what you do "in
               | response" to an ejection isn't actually in response, it's
               | about planning ahead. outside of a wartime situation
               | where factors beyond your control tell you where you'll
               | be flying, don't be in a place where it would be
               | dangerous to bail if you think you might have to bail.
        
               | jvanderbot wrote:
               | I'll do my best. So you want to dedicate probably the
               | rest of your career to automated diagnosis and recovery
               | from crash conditions after ejection? Just so we can say
               | we did a reasonable thing? Oh just the one case where the
               | pilot rejects during level controlled flight you're
               | saying we should be careful to let it continue on same
               | course and speed? And if it's slightly changing course
               | speed or altitude? Did we want to level out or continue
               | the climb and turn? Do we attempt to maintain rate of
               | climb even if it means throttling up? Descent?
               | 
               | The whole thing is so wildly ambiguous and niche that
               | it's a black hole. When a pilot ejects the controller is
               | gone. The controls are slack and it's just physics until
               | fire.
        
               | justsomehnguy wrote:
               | To add:
               | 
               | If the flight happens in the Grand Canyon?
               | 
               | A tightly populated Grand Canyon?
               | 
               | Tightly populated, divided by multiple not-so-friendly
               | nations? Which have nuclear weapons?
        
               | Merad wrote:
               | I dunno, the current approach seems quite reasonable. In
               | the grand scheme of things the overwhelming majority of
               | the Earth's surface is empty space where a plane crash is
               | unlikely to cause much damage. You also have the
               | complication that military pilots usually try to make
               | sure their plane will crash in a "safe" area before they
               | eject - many have died because they waited too long to
               | eject trying to avoid a populated area. Giving the plan a
               | mind of its own after they pull the handle would be
               | unlikely to go over very well. I believe the scenario of
               | a pilot ejecting from a perfectly good plane that keeps
               | flying for more than a few seconds has only happened
               | perhaps a dozen times in the entire history of aviation?
               | Not really worth worrying about.
        
               | dragonwriter wrote:
               | > The plane, after I eject, should do something
               | reasonable.
               | 
               | There's a good chance that it can't, and its not
               | impossible that trying to do something reasonable
               | combined with damage that led to and/or resulted from
               | ejection could make things worse.
               | 
               | > starts broadcasting a mayday?
               | 
               | Great idea for peacetime over the homeland, maybe a
               | _very_ bad idea for military operations over contested or
               | enemy territory.
               | 
               | > crashes into the nearest large body of water?
               | 
               | > attempts to fly itself back to base (we have the
               | technology)?
               | 
               | If either of these are useful in a nontrivial share of
               | ejections (except perhaps the former in conditions where
               | it takes no special effort), then there is a serious
               | problem with the training of the people pulling ejection
               | handles and that needs to be fixed, rendering the action
               | not valuable.
               | 
               | > Why would it be controversial to say "Look, guys, we
               | should decide what the plane does after the pilot ejects.
               | Maybe the best policy is just flying same course and
               | speed until fuel exhaustion, but we should choose this
               | policy, not default into it without consideration."
               | 
               | Because ejection is an action chosen when you can no
               | longer meaningfully say what the plane does in any
               | significant way. That's the whole purpose. If it it is
               | useful to address this question _then you have a bigger
               | problem that you need to urgently fix first_.
        
               | throwaway173738 wrote:
               | Yeah it's equivalent to asking what a car should do after
               | a 60mph collision. Probably sit still and try not to
               | catch fire.
        
               | appreciatorBus wrote:
               | If we had the technology to fly the aircraft back to base
               | then there wouldn't be an ejection handle, there would be
               | a "fly the aircraft back to base" button.
        
           | jon-wood wrote:
           | Even in that case this is a military aircraft, one of the
           | most highly prized secrets many nations hold. Probably the
           | most reasonable response to an ejection the aircraft could
           | take is to nose dive into the ground and slam the throttle to
           | afterburners, which isn't far of what will happen naturally
           | as soon as the pilot isn't asking the plane to stay straight
           | and level.
        
         | RobotToaster wrote:
         | Planes that keep flying after an ejection do happen
         | 
         | https://en.wikipedia.org/wiki/Cornfield_Bomber
         | 
         | https://www.marinecorpstimes.com/news/your-marine-corps/2024...
        
           | tekla wrote:
           | These are not equivalent.
           | 
           | The first one, the Airplane was in an uncontrolled spin, the
           | ejection happened to fix it.
        
         | whycome wrote:
         | This is silly. And not true. There is no "would" other than
         | your own prediction. What if the pilot deliberately wanted to
         | crash the plane but not do it intentionally?
        
         | bravoetch wrote:
         | Unlikely, you say? That's why it's being discussed. Like, if a
         | pilot ejected because the engines stop, but the control
         | surfaces still work... Maybe the plane avoids a kindergarten on
         | its dive to the earth-sky interface.
        
           | LorenPechtel wrote:
           | The plane doesn't know this. But this is part of why the ACES
           | II seats are capable of zero-zero ejections. One of the
           | scenarios in mind is allowing the pilot to steer for the open
           | space without making it a suicide mission.
        
         | tgsovlerkhgsel wrote:
         | At the very least, something to keep the flight path
         | predictable would make sense, to give the pilot a chance to
         | point the plane at a "safe" area to crash in before pulling the
         | lever. I remember reading several stories of pilots taking
         | "where is the plane going to crash" into account for their
         | ejection decision or last moves before ejection.
         | 
         | There were also several incidents where a pilot ejected because
         | the plane was somewhat controllable but it was clear it
         | couldn't be _landed_ safely. At least one of them where they
         | had tens of minutes of controlled flight before ejecting (they
         | flew it over the ocean to minimize the risk of collateral
         | damage).
        
         | the__alchemist wrote:
         | Not necessarily. For example, you'd eject if unable to get as
         | safe landing configuration.
        
       | cosmicgadget wrote:
       | Thinking about this one
       | (https://theaviationist.com/2025/02/12/ea-18g-growler-
       | crashes...), it seems like after ejection you'd want the plane to
       | lawn dart whenever possible. It allows the pilot to know if it is
       | an okay place to ditch and it minimizes the reverse engineering
       | risk.
       | 
       | I imagine there is a good reason this isn't the way things are
       | though.
        
       | Spooky23 wrote:
       | This thread is a great example of how engineers by nature are
       | tempted to add complexity to nearly any scenario.
        
       | bragr wrote:
       | The analysis and conclusions of the responders here (2018) seem
       | pretty invalidated by the 2024 F-35B ejection incident. Maybe
       | more thought should be put into what the autopilot should do?
       | 
       | https://www.marinecorpstimes.com/news/your-marine-corps/2024...
        
         | appreciatorBus wrote:
         | A single incident doesn't invalidate anything. No one has
         | argued that it's impossible for a fighter to fly on for a
         | significant period of time in good shape after a pilot ejects,
         | or the pilot has never ejected mistakenly out of an air worthy
         | aircraft. Rather the argument is that this is a vanishing rare
         | occurrence and the complexities of trying to implement an
         | unmanned auto pilot of a potentially damaged aircraft are
         | probably not worth the handful of times it might be used. One
         | incident doesn't disprove that it's vanishingly rare.
        
       | ffb7c5 wrote:
       | I think we should make an API call to an LLM with the current GPS
       | location to decide what to do, bonus points if we can mount a
       | forward facing camera and upload the picture as well
        
       | jcul wrote:
       | > The only victim of the accident was Wim Delaere, a computer
       | science student reported to have been either 18 or 19 years
       | old.[4][5][1] He was sleeping alone after celebrating the end of
       | his university exams the previous day when the MiG crashed and
       | killed him at 10:30 am. His mother and brother were shopping for
       | groceries in Kortrijk, and his father was working in Ypres.[4]
       | 
       | From the linked Wikipedia article on one of the answers.
       | 
       | What an unlucky kid.
       | 
       | https://en.wikipedia.org/wiki/1989_Belgium_MiG-23_crash
        
         | deadbabe wrote:
         | His death later inspired the movie Donnie Darko, about a kid in
         | 1988 who is killed by a jet engine in similar circumstances.
        
           | PNewling wrote:
           | I can't find anything that would suggest this is true, but I
           | would love if you were able to link me something proving me
           | wrong, as I loved that movie growing up and like reading
           | about it still.
        
       | ternus wrote:
       | I was astonished at the claim that pilots ejecting can lose half
       | an inch of height due to spinal compression induced by the
       | G-forces. Claims seem to be borne out:
       | 
       | https://www.forcesnews.com/news/can-ejecting-aircraft-make-p...
       | https://www.quora.com/Do-pilots-lose-height-when-they-eject-...
        
         | ern wrote:
         | Making sure that the consequences are well-publicized also has
         | the useful side-effect of making sure that ejection is regarded
         | as a last resort.
        
           | varjag wrote:
           | You don't really get a rando who would be unaware of that in
           | the pilot seat of a fighter jet very often.
        
       | dang wrote:
       | Discussed at the time (of the article):
       | 
       |  _If a pilot ejects, what is the autopilot programmed to do?_ -
       | https://news.ycombinator.com/item?id=17391550 - June 2018 (76
       | comments)
        
       | jayd16 wrote:
       | I wonder if dumping the fuel ASAP is a good idea or bad idea.
        
       ___________________________________________________________________
       (page generated 2025-10-31 23:01 UTC)