Piper Announces First GA Aircraft with Autoland Capability
piper.com
piper.com
There is a thunderstorm between the airport and the runway
The nearest runway is closed due to a blizzard
There is a mountain between the normal 3-degree glide-slope and the runway, an "offset approach"
There is a 40-knot crosswind
You have an engine out
Because it's an emergency feature engaged by a passenger or automatically, it probably can't manage all complex situations. It may be just good weather emergency landing. Check fuel and wind, steer to nearest emergency airport and land. Best effort is better than no effort and dying.
> There is a mountain between the normal 3-degree glide-slope and the runway, an "offset approach"
I think it can land only to those airports that are it's database with landing approaches programmed it.
Cruise missiles have had similar technology for a long time.
I try to make my landings a little less eventful.
During those first few seconds after I hit the Autoland button, the system went through a series of complex calculations and decision-making processes to determine the nearest suitable runway based on runway length, width, and surface; fuel remaining; crosswind component; terrain; obstacles; and general weather information. The system requires an RNAV approach, but beyond that, the runway and weather criteria can be decided by the airframe manufacturer.
The system even forecasts its own weather if the nearest suitable runway is a significant distance away, long enough that the current ADS-B or SiriusXM weather may not be valid. It uses the latest weather trend information, for example, to determine if a thunderstorm might move into the runway environment where it intends to land. It will route the airplane around thunderstorms as well as terrain and obstacles, all of which it gets from its internal databases. If en route to a runway it determines, because of changing weather conditions, that another runway is closer or more suitable, it will change its destination. It can even estimate changing barometric conditions and adjust the altimeter—using algorithms. Garmin engineers say the calculated barometric readings are within 0.01 inches of mercury of actual ambient conditions.
Even then, it's problematic: https://en.wikipedia.org/wiki/Iran%E2%80%93U.S._RQ-170_incid...
Cruise missiles aren't trying to survive the landing.
Until GPS guidance came into use they were relatively crude terrain followers requiring a preloaded flight plan over a known terrain profile with sufficient bumpiness to protect against IMU drift. Nothing at all like this technology.
Yeah but an average human couldn't, and thankfully still cannot buy one. This is feature on a (and more to come) civilian aircraft, two very different things.
And if the aircraft is low on fuel or isn't making enough power to make that runway... Well, the usual happens - a forced landing, aka a crash landing.
I don't think this is meant to be used as a cop-out by the pilot. "My engine quit! You deal with this, computer!"
Closure: certainly a problem.
Mountain: can use worldwide topographic information in its database to route around. See AVweb
Crosswind: mentions getting local weather.
Engine: Autland cannot help. But on the Cirrus you'd pull the CAPS cord and float down.
At least a non-pilot co-passenger might want to try their luck gliding to open area and then pulling it.
A fall of 10 feet puts you at ~17 mph when you hit. 20mph is a good approximation.
You’ll go wherever the wind pulls you. And might hit the side of something faster than 20mph before hitting the ground.
And hitting a crowd at 20mph.
Well of course they do.
CAPS does a pretty good job putting an upper bound on the worst case result. They don't care if you screw your plane up because that's a known cost and you'll pay it back through premiums if you keep flying.
Dead sticking it has a very good chance of being no more expensive than a normal landing if you find a runway and not all that expensive if you find a good field or empty highway (i.e. a big towing bill) so of course most pilots would prefer to dead stick. But this comes with a small chance of you crashing into a daycare and maxing out your policy. The insurers do not want this.
Most decent pilots with good situational will be able to make the correct call between CAPs and dead sticking it. They're telling you to go CAPS all the time because they have to deal with idiot pilots too. It's the same reason every drivers ed program emphasizes brakes, brakes, brakes, despite the steering wheel being a far better way to avoid many problems. They don't trust you to make the call.
[citation needed]I know of a skydive place that does or at least used to regularly (like once or twice a week) dead stick it in a C182 because they would fly with the bare minimum of fuel (when you're flying up and down all day even a little extra weight costs a lot of money).
> > There are very few emergencies (in my opinion) in which parachute deployment would be the appropriate response unless there was an extremely unfortunate set of circumstances which led to the airplane being unable to fly or glide (like a wing snapping off)
> I've been a Cirrus instructor for 15 years. This attitude is completely wrong and is exactly what has led more people to be killed than needed to be, especially when Cirruses first came to the market and the chute was a new thing in production aircraft. We TEACH, intentionally, people to ask FIRST, "do I need to use the chute," and the answer is never "no." It's either, "Yes," or "not yet."
> There have been pilots - plenty of them - that had every opportunity to use the chute but kept gliding, or spinning, or whatever, and now they and their passengers are dead.
[...]
> Lose the engine - most of the time, pull the chute. Yes, even better than landing in a nice open field. The chute works. If I screw up the landing, or the field has ditches, furrows, irrigation, cows, power lines, fences, rocks, or who knows what else, I might just be screwed setting a plane down at 65-70kts. But not if I float down into it protected by my roll cage, high-g seats, air bags, etc.
[...]
Chute pulls are almost 100% survivable. Landing in a field is a much lower percentage. Just hitting nearly invisible power lines can slice a plane in half. You also have a fire danger when landing off airport. And many dead engine landings result in landing short of the field.
You aren’t going to get ridicule from the Cirrus community for pulling the chute and even if you did, it’s less ridicule than crash landing into a field that looked suitable from 2000 feet up.
It's almost impossible to tell what the situation is on the ground until it's too late--an unnoticed ditch can kill you.
Which would suggest that if you need to be vectored to a runway, the aircraft can negotiate that with ATC automatically (and perhaps ATC can advise which is the most appropriate diversion?) This in the situation where the pilot is incapacitated, but the aircraft isn't in an immediate risk of crashing (e.g. you're cruising, but there's bad weather immediately below you).
It should also be able to use stuff like D-ATIS to get airport conditions.
That seems optimistic though, perhaps it just announces its intentions and hopes ATC clears a path?
I strongly suspect it just announces itself.
"N123FOO DECLARING EMERGENCY; PILOT INCAPACITATED, AUTOLAND ENGAGED; CLEAR RUNWAY 21L; 12 MILES OUT" sort of stuff.
Remote truckers in an office, similar to a UAV drone pilot.
This is not an option in a plane. If it was, this technology wouldn't be needed - if the pilot is incapacitated just brake to a halt and step out of the plane...
Possible scenario: if pilots are already sitting in a simulator for currency training / testing, it may be possible to have the sim be 'reset' to a live feed in an emergency.
You wouldn't have them hang around 'just in case', but rather use people who are already present for some other reason.
Another possibility: pilots are hours-limited for a given time period. If you have pilots that have hit that limit, some may be willing to spend time in a ready-room for a nominal bonus.
(It'd probably also be significantly more expensive, due to the need for a subscription.)
But I could imagine a company contemplating building the remote-pilot-for-emergencies service might be scared off by potential liability. Families of pilots killed in GA accidents routinely sue aircraft and engine manufacturers even when the FAA determines the cause to be pilot error.
Starting at 1:30, "NOVEMBER SIX HOTEL LIMA, POSSIBLE PILOT INCAPACITATION, SIXTEEN MILES WEST OF KILO CHARLIE LIMA--" at this point the voiceover talks over the announcement, and I have trouble making it out, but I'm pretty sure I hear a "...LANDING ON RUNWAY..." in there as well. For those of you that don't know pilot jargon, that's the plane's registration number (N6HL), the situation, location relative to the nearest airport (KCL-something), and what I'm pretty is a declaration of intent to land on a specified runway.
So, your guess is correct, what the system does is announce its status and intentions and let ATC clear the way.
Other comments have speculated about uncontrolled fields, but I think that's less likely to be an issue. Per the demo video, the auto-land feature requires a GPS approach with altitude guidance published. If a field can't afford a staffed tower, would it be able to afford to have a published GPS approach with altitude guidance? My intuition says no, but as I'm not a pilot, I could be wrong about that. On possible scenario I could see is that the tower isn't 24x7, and so the field is uncontrolled at night. Perhaps such a field would have a GPS approach published, and if the auto-land selected that field, it would be up to any pilots in the area to be aware of the announcement and get out of the way. AFAIK, having one of your radios monitoring the emergency band is considered best practice, so other pilots in the area should be able to hear the emergency announcement and react appropriately.
But at this point, I'm speculating as well. For all I know, the auto land only selects airports known to have a manned tower at the time of activation.
The accident plane will be constantly reminding people of its status on this frequency when there isn't a tower. Other pilots will be talking in response to each other but this system can't understand what they're saying. So it will just explain periodically what's going on. Something like:
"Springfield Traffic, November Six Zero Hotel Lima with Emergency Possible Pilot Incapacitation. Final Runway One Six. Full Stop. Runway Will Close After Emergency."
I can definitely imagine, especially when this technology is new, other pilots trying to ask questions
"Zero Hotel Lima, did you say runway will close? I was planning to land there after you. Do you know when it will re-open?"
The system can't understand their question and won't answer, but their training should teach them De Niro's line from Ronin. "When there is any doubt, there is no doubt". Something is wrong, they should divert and leave the emergency plane to do its thing.
The vast majority of reasonable places to land something small like these $2-3M planes will not have a tower at all, and for some of the rest the tower is only part time, but your emergency may not be, so it is usually going to be talking on CTAF and other pilots will just have to make new plans to work around the emergency.
You handle those things manually by setting proper inputs to the normal non-emergency autopilot or flying without autopilot like you would in any other aircraft equipped with an autopilot that is capable of landing. I assume "one off" approaches are a non issue because they are using a database of supported airports with the necessary approach details since they change slowly and the list is short enough. That's traditionally how this kind of thing is handled.
This is not a replacement for a pilot and is not portrayed as such.
I don't know if this can deal with the engine out, but in theory there isn't any reason it couldn't. There is an IPad app called "Xavion". With an engine out, Xavion will calculate an approach for you follow to a suitable airport, based on the aircraft's best glide speed + required runway length. Interestingly it doesn't use the published best glide speed and stopping distances, the app uses your flight history to figure out the aircraft's actual best glide and stopping distance.
Would the costs for short hops be comparable to a commercial airplane ticket? Because it would be awesome to charter small flights from an app to jump to a nearby city.
Bear in mind even in takeoff and landing scenarios, unlike a public roadway, the airplane has pretty exclusive use of the runway as well.
Pilots are very unlikely the primary cost of aircraft trips. Things like jet fuel are probably much more of the cost driver.
The environment has shown quite clearly that we don't want to. We'd probably already have it if we did.
The safety design of aviation relies on having multiple highly competent people in the loop on every flight and exercising their judgement on dozens of things. And that's for the best-case scenario, they also have to be capable of choosing and executing fallback plans when part of the overall system (the plane, air traffic, radio navigation, flying conditions, ground conditions, etc...) degrades or enters an unexpected state.
Mandatory bad car analogy: a driver will often prefer a faster response over spending time to check the rear view mirror before slamming the brakes, a computer would never have reasons to do so, they can check without meaningful delay.
And maybe catching/handling them properly would involve more sensors, when people really just want a well-trained human who is able to improvise in the cockpit.
It would be difficult code a machine to find a good landing spot in extremely variable terrain. For instance, I saw a video of a plane safely crash-landing landing on a four-lane highway with traffic. It would be impractical to code something to handle that very rare situation.
Even experience cannot be much of a factor on top of that, as very few pilots will ever experience a scenario like that outside the simulator. My layman's suspicion, honed in the MCAS debates, tells me that the only other thing experience and human "general intelligence and gut feeling" can add to the equation is educated guesswork regarding which data to distrust when things just don't add up consistently due to a malfunction.
> V_1 is the critical engine failure recognition speed or takeoff decision speed. It is the speed above which the takeoff will continue even if an engine fails or another problem occurs, such as a blown tire.[9]
* https://en.wikipedia.org/wiki/V_speeds
The plane has to keep going until V_2 (or V_2_min).
One thing that no doubt helped with Sullenberger was he was a hobby glider pilot where that kind of situation comes up regularly if in a much less dangerous way as gliders are slow and light.
(I don't mean to make it sound as I think it's easy to solve, but the basic strategy to solving it is simple :)
It's a problem of trust, and regulation, not a technical one.
I doubt the cost of a pilot's time alone is a big enough factor to meaningfully change the number of people with access to chartered flights.
The other costs (hours of plane time, fuel, ramp fee, parking fees, maintenance) probably dwarf pilot time as a percentage of total cost. And you're either pushing the absolute unreliability of aviation [0] to the customers who are now trapped in a different city, or they're paying even more to the app to have backup plans ready to roll.
[0] Commercial aviation deals at a scale that can absorb flights that don't keep their plan. You go to the counter, get rebooked, and angrily eat some shitty fast food. When something goes wrong on a private flight, you're pacing around on the tarmac at 1 am, 3 states away from home and have to be at work in the morning.
[1] https://lilium.com/newsroom-detail/taking-to-the-skies-fligh...
To be considered safe, an auto-fly needs to be able to gracefully degrade as the subsystems in which it depends start to fail. Human pilots are trained to handle this by compensating for system failures in a variety of ways, including compensation (eg in a split flap scenario) and disabling (turning off a misbehaving autopilot). Ultimately, a human pilot can always turn off all the automation and fly the plane by looking out the window (or, in poor conditions, using gyro powered “steam” gauges).
In the case that the pilot is disabled, then auto-land is a great option, even though it’s reliant on so many subsystems. But for an always-on auto-fly, the requirement to gracefully degrade, and degrade, and degrade, and still operate at the level of a human pilot using the “Mark 1 eyeball” is a tough one which hasn’t yet been met (as far as I know).
There’s mention in this discussion of fully automated drones. But, those are basically expendable if there’s a serious system failure, since there are no souls on board.
What happens then they do?
A couple highlights from the Records section:
> On 24 April 2001, a Global Hawk flew non-stop from Edwards AFB to RAAF Base Edinburgh in Australia, making history by being the first pilotless aircraft to cross the Pacific Ocean. The flight took 22 hours, and set a world record for absolute distance flown by a UAV, 13,219.86 kilometers (8,214.44 mi).[97]
> On 22 March 2008, a Global Hawk set the endurance record for full-scale, operational uncrewed aircraft UAVs by flying for 33.1 hours at altitudes up to 60,000 feet over Edwards AFB.[98]
Unlike Predator/Reaper/etc. drones, the Global Hawk is autonomous, there is no pilot sitting in a faux-cockpit stateside.
I'd be interested to know if it declares an emergency, squawks 7700 and controllers are expected to give it a priority and right of way?
What I’m trying to say is that the auto land functionality is not a feat, it’s like self parking cars.. been around for 20 years just now cost efficient and safe enough for the masses to use. In a boeing 787, the pilot still needs to squawk 7777 and they should try comms, don’t know why a GA plane would be different..
The system being discussed is for when you no longer have a pilot.
This is much more impressive, not the least of which because the system itself figures out the best place to land (taking into account operational constraints), how to get there (taking into account weather avoidance), and how to be at the right speed, altitude, configuration, etc. to perform the final approach to landing.
So to use your analogy, getting a bit closer to a self-driving car than just self parking!
Regardless, the machine will be broadcasting its intentions and again, there's no arguing with a machine other pilots will just have to accept that it's intending to land and stop on the runway putting it out of use until further notice.
Since the machine can't see lights, I assume it doesn't care whether they are functioning. So in the dark it's likely to land at some small airfield which is actually closed, slightly scary for the passengers to just come to a halt in darkness far from anywhere but arguably safer since nobody else is going to make the error of landing when the runway remains occupied.
No need:
> [Autoland] would update and broadcast that message every 30 seconds—listening to make sure that it didn’t transmit over any other radio calls. Once near the Class D airspace of the tower, it would have changed one of the radios to the tower frequency and kept the other on the emergency frequency.
Then, once landed, it will brake, and then broadcast that the runway is inop due to airplane on the runway.
Quite amazing.
https://www.aopa.org/News-and-Media/All-News/2020/January/Pi...
"The routing algorithm determines a path to the Final Approach Fix (FAF) of the chosen published approach that avoids terrain, obstacles, and significant weather."
That handles all of your contingencies except an engine failure. If you have an engine failure and a pilot who can't dead-stick a landing you're probably screwed. But you're still no worse off in that case than you would have been without autoland.
We can require more training for GA, but the training burden is already pretty high, and the GA industry is still shrinking last time I checked. Boosting safety requirements in the form of additional training for GA may save a few lives, but also those gains could be simply because there will be fewer hobby pilots.
The fact is that GA planes falling out of the sky just doesn't kill very many people. It obviously feels terrible to the families involved in a crash, but relative to automotive deaths, lung cancer, murders, etc, it is a tiny tiny blip in the numbers. Obviously being a GA pilot boosts your chance of dying in a small airplane crash, in the same way that learning to ride a motorcycle boosts your chance of dying in a motorcycle crash. The alternative of killing off recreational flying, is worse.
I seem to remember reading like 4 or more years ago, that there was an experimental autoland system actually powered by that software, and could autoland in a dead stick situation.
EDIT: Found the experimental engine-out auto-land system based on the same technology as xavion: It is called the VP-400 by Vertical Power. I'm not sure if it ever became a purchasable product.
Represent each airport as a node in a graph with all relevant data (distance, weather, etc.) ... Traverse the graph and find the shortest path.
https://en.wikipedia.org/wiki/Dijkstra%27s_algorithm
Easier said than done I'm sure ;)
The word "just" in their post may be prickly. Also, it's not really a graph search as stated, just a "choose the airport with the lowest weight". That, in theory, chooses the airport to land at, but not... the rest of it.
Yes, choosing a route would be essentially a graph search, with weather at intermediate nodes adding to edge weights, etc, but even after you've computed a target path, you're going to be continually refining that path (e.g. the thunderstorm moves in a way you didn't predict). And... flying the aircraft itself.
Autopilot in straight & level flight isn't too big of a deal. Autopilot tracking a path is doable (including ascent/decent). Auto-throttle is common on larger airliners but not in GA aircraft. Automatically configuring landing gear and flaps, as appropriate for weather conditions, is (afaik) not done automatically even in commercial airliners. And doing all of this, reliably, with the assumption that the pilot is incapacitated... that's friggin' cool!
At a very naive level, for example: typed language vs dynamic; formal verification of the code or not (or to what extent); what kind of QA? (can't even imagine how to test this "in production" given the amount of edge cases like the ones you describe; hw and sw redundancy; to name a few.
I have zero experience with airplane software but I find it fascinating considering how safe flying is.
Anyway, I'm sure there used to be a blog for the Dreamliner test team, but I can't find it now, so here's this instead: https://www.wired.com/2010/03/a-look-inside-the-brains-of-bo...
And specifically for code, look here: https://ldra.com/aerospace-defence/standards/misra-cc/
and (JSF, not commercial) https://www.f35.com/about/life-cycle/software
You are not approaching this from the correct angle. "dynamic" and "static" types is a red-herring for safety. Yes, when we are rushing some feature in our day jobs we can try to call functions with their wrong types. But these are not critical systems.
If you DO have a critical system, you are going to have to invest _massive_ engineering resources. You are going to exercise all code paths. If you get a "type" error, you haven't done your job correctly. Because this was a code path not exercised before, and that should not happen. The additional type checks at compile time are a drop in the bucket compared to all the testing you have to do.
That said, most such applications will be very traditional. C or C++, maybe ADA. Despite having types, the C family you give you a whole lot more failure scenarios. For instance, it's very easy to corrupt your entire process memory in C, that is something that not many "dynamic" languages will let you do. And yet people build mission critical systems with them.
That's not the only possible answer though. For an interesting autonomous system, please look into NASA's Deep Space 1, in particular the Remote Agent. Written in Lisp, it had spacecraft controls for a few days. It could self-diagnose and correct issues. For a long time, it was the most advanced piece of software to ever control a spacecraft. And Lisp is "dynamic" typed (but strongly typed)
Now, they do test in "production", but it's more like they have lots of staging environments. Some of them may be simulators, some of them may be replicas. At the end of the day, this is no different from any other software: it has inputs (airspeed, altitude, heading, roll, pitch, yaw, servo positions, charts, weather radar, etc) and outputs (controlling the servos, changing radio frequencies, etc). You can simulate some or all of them.
* https://www.avweb.com/aviation-news/garmin-announces-emergen...
* https://www.youtube.com/watch?v=d-ruFmgTpqA
* https://www.ainonline.com/aviation-news/general-aviation/201...
Also Cirrus:
^ I saw this in my feed before this HN article.
Looks like Garmin coordinated the announcement.
This is an ipad app that calculates a best glidepath to the best runway and guides you to the runway safely. It's intended for emergency situations when you lose an engine and quickly have to decide on an airport/runway. However, it sort of snowballed into an app with a lot of situational awareness features.
But he actually experimented with hooking that up to a plane at some point to create a self landing system: https://www.x-plane.com/2014/07/xavion-brings-a-real-world-a.... Not sure if he ever shipped that feature. I imagine there might be some certification hassle :-).
Of course I don't think it was as fully integrated, so no automatic radio callouts or anything.
I'm not sure if the product actually ever became available.
HOWEVER: If you have a TruTrak autopilot, and an an iLevil ADS-B receiver, then Xavion has a super experimental "auto" button that will actually control the autopilot, to attempt to automatically fly the calculated course.
Not to be confused with CAT II/III autoland used in normal operations.
From Wikipedia: "The feature is activated by a guarded red button on Garmin G3000 NX avionics, evaluating winds, weather and fuel reserves to select a suitable diversion airport and taking over the aircraft controls to land, it advises the ATC and displays instructions to occupants."
From the Garmin marketing site [1]: "Garmin Autoland. It takes complete control of the flight to land the airplane in an emergency where the pilot is unable to fly1."
I'm not exactly sure if this is the correct reference for CAT II/III autoland, but seems like the purpose of CAT II/III autoland is to land and aircraft when the pilot would otherwise have a hard time manually landing due to weather conditions (ie. limited visibility) [2]. But if someone has a more accurate description, please correct me!
[0] https://en.wikipedia.org/wiki/Autoland#Emergency_autoland
[1] https://www.garmin.com/en-US/autonomi/
[2] https://en.wikipedia.org/wiki/Instrument_landing_system#ILS_...
Edit: Just to clarify the reason for posting this versus just relying on the text in the press release, I was under the impression the announcement was for some fully-automated flight control system (ie. take off and landing without a pilot on board, a la self-driving cars) based on the press release not mentioning why the system would be activated.
CAT IIIc is interesting because there is no decision height and no runway visual range limitation. Pilots can approach and land completely blind without flying with the stick at all. https://www.youtube.com/watch?v=gthcuZV8agI
Big airports with CAT IIIc can stay open for CAT IIIc available aircraft in all visibility conditions.
This Garmin G3000-based system primary uses GPS/WAAS (and perhaps VORs) to get location data, so accuracy may not be with-in necessary tolerances for day-to-day use, but it's better than nothing for emergencies.
Cat II ILS can go down to 100 and 1200RVR (100’ bases and 1200’ of runway visual range)
Cat IIIc can go down to 0/0.
200 and 1/2 is perfectly usable for most day to day weather and is certainly good enough for unexpectedly incapacitated pilot.
The autopilot features available to GA and experimental aircraft are quite impressive for holding things at a specific ask in various weather conditions. Not uncommon to have the autopilot active when flying in IFR conditions to help with some of the workload.
Assuming the plane has GPS, a correct altimeter setting, and airspeed - between the instruments you get a reasonable amount of cross checking for accuracy and enough information to calculate your vector. If you actually fly/land at the speeds recommended, most of the time landing will be pretty uneventful.
Many of the tablet/phone based GPS only solutions will give a pretty solid estimation of engine out glide distance too. You (should) pitch for 'best glide' speeds and in combination with the airport data bank with the GPS unit, know where you can get to. I'll push the 'direct to' nearest button and try for it were it in glide range. This sounds like it just connected all the automation dots.
A good landing is where everyone leaves the plane safely. A great landing allows the plane to be used again. I suspect this is going for a good landing... which is fantastic news for the passenger.
A typo but that slip created a great brand name for a rocket-propelled parachute the "ParaShoot".
https://en.wikipedia.org/wiki/High-altitude_military_parachu...
There is also TALO, which stands for Tactical Air Landing Operation
https://cirrusaircraft.com/cirrus-aircraft-revolutionizes-pa...
I'm assuming GA means generally available?
And I definitely see a future version that could handle engine failures. If you can count on auto land to flawlessly glide to the nearest airport, who needs a twin engine airplane? Single engine airplanes are significantly cheaper to operate than twins.
Not to mention we're still burning lead in most propeller aircraft as well.
Fortunately this isn't a very popular way of getting around.
Would a full on private jet be way worse?
Would these planes carry a fair few people than a car? If so what’s it like taking passenger numbers into account?
I guess super-rich folk probably would be happy to fly just themselves in a big private jet and also be travelling waaaay further than they ever would in a car so it’s probably a silly comparison anyway
To compare to a car, just using vehicle efficiency (not per passenger) a typical cessna is in the realm of a full size SUV in terms of carbon footprint. Both are about 10-15 mpg. The aircraft in the article is far worse but only because it's rather large and turbine powered. Optimized designs like the Long-EZ can easily hit in excess of 30mpg cruising at 160mph.
As I mentioned in a previous HN item about ILS, these autoland systems don't get you off the runway. For a grave emergency (pilot dropped dead in flight with their spouse and two kids onboard) that's forgiveable, and this mentions that the product begins broadcasting that the runway chosen is now unusable (because your plane is now parked on it) but this is why Cat IIIc doesn't magically allow JFK to function normally in a blizzard. When the pilots still can't see the runway even after their plane lands on it, that's not useful for routine passenger flights.
I wonder how this will change when the L2, L5, and L1C signals start becoming used from the GPS III satellites.
Also, in the US, non-GPS/Navstar signals cannot officially be used for anything, though I think Galileo was recently (finally) approved by the FCC:
* https://www.gsa.europa.eu/newsroom/news/fcc-approves-use-gal...
(The issue was something like: once officially approved, the frequencies in questions cannot be used for anything else, and are protected.)
But a lot of folks now have GPS, especially because of the ADS mandate, and so while most probably aren't IFR-rated, I'm guessing if you're flying a turboprop or Cirrus jet, they probably are.
It does seem like a lot of airports have RNAV only and if they have an ILS they also have an RNAV.
One mnenonic is GRABCARD
Generator Radios Attitude indicator Ball Clock Adjustable altimeter Rate of turn indicator Directional gyro
Or you can just pull the throttle to idle, and trim for Vx, which is the "descend through clouds" emergency procedure.
I would love to know if that's not true because that would be a huge leap in off-line speech processing and recognition.
How does it know? Also a weird submission to be at the top of HN
Seems like the usual HN fare. Autonomous control of vehicles is a very common topic here.
The plane could sound an alarm for you to check in if it thinks you're not conscious.
Edit: More details at https://www.ainonline.com/aviation-news/general-aviation/201...
It pretty much works like you'd imagine, with the addition that a passenger could engage it if required.
>In the M600 Autoland application, the system activates at 18,000 feet if the autopilot is engaged and the pilot doesn’t interact with the avionics in a 15-minute period. At higher altitudes, the engagement period is shorter. Autoland also tries to alert the pilot with repeated chiming sounds and asking, “Are you alert” before engaging.
Know the nearest airport, or know if the pilot is incapacitated?
1) GPS and 2) it looks like someone in the aircraft has to press a button.
Whether it's a train, a boat or in this case an aeroplane, an aware and capable operator will periodically fiddle with things a little bit, change the heading slightly, tweak engine power a bit, any little change shows you're paying attention still, no need for the machine to intervene. If that doesn't happen for a sustained period, the machine prods them. "Hey are you awake? Let me know you're OK" or moral equivalent. If they still do nothing after a further period, engage emergency systems. On a boat that means summoning the rest of the crew to come see what's wrong on the bridge, on a train it means you just bring the train to a halt and let a remote signaller communicate with anybody else on board, on a plane this system proposes to try landing.
Also a passenger can just (be taught how to) activate it.
Even after touching down, the aircraft this is built for have computer control for the brakes and full rudder authority, so it can do a pretty good approximation of a proper short field landing based on GPS and wheel speed sensors. Do you want to add computer controlled brakes and wheel speed sensors to your tin can? No? Then we have to assume the plan is to just roll to a halt and that cuts our list of viable landing sites and hurts survivability considerably.
It's a market with higher certification requirements, but much smaller unit volume than boats and cars, so much more certification, R&D, admin cost has to be covered by each unit sold.
See eg. https://dsm.forecastinternational.com/wordpress/2019/02/26/2...
It's probably like having a parachute on the plane: just a great marketing sales point.
https://www.ncbi.nlm.nih.gov/pubmed/28539144
There are already plenty of airplanes that have the ability to land themselves under specific conditions. There's no whole-new technology here. Even if it was brand new, I'm not sure how that would be less safe.
Once pilots were properly trained, the parachutes started saving lives. It was the same story when ejector seats were introduced.
The emergency autoland is not intended to be used by the pilot, it is for activation by a passenger in single pilot aircraft when the pilot is incapacitated. So their is almost nothing to train. Pilots just need to add it to the passenger briefing.
Net safety will probably increase because most people will never have that use case though.