Special Report: Seeking lessons for eVTOL pilot training in the F-35

This essential reporting from The Air Current is now available for all to enjoy courtesy of the General Aviation Manufacturers Association (GAMA).

I’m at the controls of an F-35 flight simulator, preparing to land my virtual F-35B on the simulated deck of the USS Wasp (LHD-1) multipurpose amphibious assault ship. Thirty minutes ago, I had never seen the cockpit of a Lockheed Martin F-35 Lightning II, let alone touched the controls, but by this point I have performed a roll at 35,000 feet, flown a short take-off from the same deck where I am now about to set down, and burnt through thousands of pounds of imaginary jet fuel while puttering around in a hover.

The fact that I have done all of this without incident — and now proceed to land safely on the deck of the Wasp — is a testament to the expert instruction of my simulator operator, Robert Parlett. But it also demonstrates how simple the aircraft is to fly. Designed with augmented fly-by-wire flight controls and envelope protections, the F-35 does not require its pilots to have much feel or finesse. Flying it is a largely procedural affair, and Parlett readily talks me through the necessary procedures.

The simulator session is part of my visit to the Patuxent River F-35 Integrated Test Force (PAX ITF) at Naval Air Station Patuxent River, Maryland, where I have come to learn something about the future of electric vertical take-off and landing aircraft. The F-35B is the short take-off and vertical landing (STOVL) variant of the F-35 fifth-generation strike fighter, and it uses a unified flight control scheme that was originally tested on a modified Harrier jump jet, the Vectored-thrust Aircraft Advanced Control (VAAC) Harrier. In 2015, when California-based Joby Aviation was contemplating control schemes for its novel eVTOL aircraft, it too drew on the lessons of the VAAC Harrier in electing to use unified controls.

Related: FAA issues final airworthiness criteria for Joby eVTOL

The F-35 is a single-seat fighter, so pilots train in simulators before going solo in the real thing. Joby, counting on being able to use a similar training model, designed its weight-sensitive electric aircraft with a single pilot station, and many other eVTOL developers — including Archer Aviation, Eve Air Mobility, Lilium and Vertical Aerospace — have since followed suit. In June 2023, however, the Federal Aviation Administration published proposed rules for powered-lift pilot certification that would require these companies to develop a version of their aircraft with fully functioning dual controls — a costly exercise that would impose the greatest burden on the companies closest to certification, including Joby.

The continued existence of the F-35B is proof that it is eminently possible to train pilots for safe operations in a single-seat aircraft, even one with vertical flight characteristics that differ significantly from what they have flown before. What I sought to understand is what exactly about the aircraft and its training program have made this model a success. Is it the design of the flight controls? The quality of the simulators? The structure of the training program? The answer, I learned, is all of the above — plus one additional element that speaks to the role of the professional pilot in an era increasingly defined by automation.

Rethinking flight controls

It is a peculiarity of most pilots that the very qualities that make an aircraft difficult to fly — the twitchiness of a tailwheel plane or unstabilized helicopter — can also engender affection. And it is with affection that U.S. Marine Corps Lt. Col. Mike “Latch” Lippert recalls the McDonnell Douglas AV-8B Harrier, the jump jet whose A model was dubbed “The Widow-Maker” for its exceptionally high accident rate.

“It was a lot of fun,” said Lippert, who is now the F-35 government flight test director at PAX ITF. “It was a challenging airplane to fly, a challenging airplane to stay ahead of. But there were rules of thumb that you just had to always live by, and you had to respect the aircraft a lot.”

The Harrier, first developed by Hawker Siddeley in the 1960s, was revolutionary as the first operational attack aircraft with vertical/short take-off and landing (V/STOL) capability. In theory, this enabled the aircraft to operate from confined spaces such as improvised bases and the decks of small ships. In practice, however, the aircraft’s usefulness in some instances was limited by its poor flying qualities.

The Harrier maneuvers in low-speed flight by directing exhaust from its Rolls-Royce Pegasus turbofan engine through four synchronized rotating nozzles, and compressor bleed air through reaction control valves on the nose, wingtips and tail. Managing propulsive forces in this thrust-borne flight regime can be challenging for the pilot — who must work both a throttle and a nozzles control lever with their left hand — and additional handling challenges result from the complex interaction of propulsive and aerodynamic forces when transitioning between thrust-borne and wing-borne flight modes. Add in poor visibility, gusting winds or a pitching ship deck, and the combination can quickly exceed the pilot’s ability to maintain safe flight.

With the advancement of digital fly-by-wire technology, precision inertial sensors and electronic displays, researchers began seeking better control schemes. Initial research was performed in the 1970s and ’80s using large motion simulators at NASA Ames Research Center in California. As the Joint Strike Fighter program advanced through the 1990s and into the 2000s, two key concepts from this research were further developed and refined using two different Harriers modified with digital flight controls: the single-seat NASA V/STOL Systems Research Aircraft (VSRA) in the U.S., and the two-seat VAAC Harrier in the UK.

Related: Newly proposed FAA rules could create eVTOL pilot training hurdle

One of these concepts was translational rate command (TRC) for precision hovering and vertical landing maneuvers. Under TRC, the aircraft’s speed over the ground is proportional to control displacement. Moving a control stick a small amount to the right causes the aircraft to move slowly to the right; as the stick is moved farther to the right, the speed increases. When the stick is centered, the flight control system returns the speed to zero and the aircraft maintains a stabilized hover.

TRC as I’ve experienced it in the F-35 and Joby simulators feels like magic, because it’s very different to how flight controls function in conventional vertical-lift aircraft. In a traditional helicopter, for example, moving the cyclic stick to the right causes the entire main rotor disc to tilt to the right. This creates an acceleration proportional to stick displacement, and the helicopter will continue accelerating to the right until the aerodynamic forces balance or the pilot decides that’s fast enough and takes out some of the right control input. This type of acceleration command or progressive rate control is useful at higher airspeeds, but it creates a high workload for the pilot when maneuvering in a confined environment or trying to land on a precise spot — a problem that also afflicted the Harrier in a hover.

The other key concept was unified flight control, which is based on the premises that the transition between thrust-borne and wing-borne flight should be intuitive to the pilot and that the axis of control associated with a given inceptor should remain the same in both flight regimes. As first implemented in the VAAC Harrier, there are two inceptors for unified control: a right-hand inceptor (RHI) in the form of a sidestick, and a left-hand inceptor (LHI) that resembles a throttle. In thrust-borne flight, moving the RHI aft or forward causes the aircraft to move vertically up or down, respectively. As airspeed increases, the RHI continues to command altitude changes, but now through the more familiar mechanism of pitch attitude — pointing the nose of the aircraft up or down.

Likewise, moving the RHI left or right causes the aircraft to move left or right over the ground in a hover, and commands a left or right roll rate at higher airspeeds. Meanwhile, moving the LHI fore or aft causes the aircraft to move forward or backward over the ground in thrust-borne flight, and to speed up or slow down when flying on the wing — just as a throttle would function in a conventional airplane. The inceptors typically provide acceleration commands, but in hovering flight, the pilot can choose to activate TRC for precision maneuvering.

The results of the VAAC Harrier tests were decisive. Unified control proved to be so intuitive that the program even allowed a non-pilot journalist, Lawrence McGinty of ITV News, to fly the VAAC Harrier for a news segment. “I’m fat, I’m 50, I’m trussed up like a chicken, I can’t even drive a car, but I’m going to have a go at driving this,” McGinty in full flight gear memorably intoned from the cockpit before his flight. (And afterward: “I don’t believe it! That was — wow! How do you find sex interesting after this?”)

“Clambering out of the cockpit, I had to agree the Harriers’ replacements with the new system will be easier and safer to fly and allow pilots to concentrate on fighting rather than just flying,” McGinty reported. With unified control now implemented on the F-35B, that prediction has unquestionably come to pass.

“Like I said, [the Harrier] was a lot of fun to fly. I’ve got a certain special place in my heart for it,” Lippert said. “But there’s no comparison to the level of effort. I mean landing that airplane on a ship at night, versus landing an F-35 on a ship at night in the worst sea states — you cannot compare.”

A simulator-first development path

It is often said that the flight controls in the Joby aircraft are the same as those in the F-35B, but it’s actually more accurate to say that both were informed by the lessons of the VAAC Harrier. That’s according to Peter “Wizzer” Wilson, Joby’s director of flight standards and training, who previously served as lead test pilot for the F-35B program. At the moment, he is the only person in the world who has flown both aircraft.

“Forget the F-35 program for a minute and imagine in 2015 Joby is trying to figure out with this complex aircraft they’re putting together what would be the best control scheme,” Wilson told me. “The flight test people here were looking into various papers, they found Buddy and Jif’s paper from back in the day from the VAAC, and independent of the F-35, they decided unified would be the way to go.”

“Buddy” and “Jif” — otherwise known as James Denham and Justin Paines — are engineering test pilots who led research and development of the unified control concept and published their findings in a series of technical papers. Paines served as Joby’s chief test pilot for three years before assuming that role at Vertical Aerospace in the UK; his position at Joby is now held by Denham. Because Joby’s eVTOL shares a common ancestor with the F-35B rather than being a direct descendant of it, there are some differences in how unified control is implemented in each aircraft, but the underlying philosophy is the same.

Related: Why eVTOL sims could be a bridge to the future of flight training

“Height control is always in your right hand, and speed control is always in your left hand, and that’s the fundamental concept. In that regard we’re very similar to the scheme on the F-35B,” said Wilson (noting that the RHI technically commands flight path rate rather than pitch rate). Likewise, on each aircraft, TRC is activated with a simple button press.

Wilson said that some of the control rates are different “because the role is different … you really don’t need to be able to move a [passenger] aircraft as quickly as you can move a fighter aircraft.” And whereas the F-35B has active inceptors — meaning they have different control feels and detents at various stages of flight — Joby’s aircraft has passive inceptors for lower complexity and cost. “That means that we do not have features that change from one flight regime to the next, but again, our flight regimes are much, much simpler — we’re not trying to go out and do air combat against somebody else, so we don’t need to have those more advanced parts of the control law,” Wilson said.

The most significant difference is that the F-35B has conventional rudder pedals for yaw control, while Joby has incorporated yaw control into a twist grip on the right-hand inceptor. According to Wilson, this was done primarily to reduce the weight of a pedal assembly, since every pound counts in a small electric aircraft.

Wilson granted that moving yaw to the RHI for Joby’s aircraft has “interesting human factors connotations. Suffice to say … we are comfortable with this as a way forward.” Technically, a fly-by-wire aircraft doesn’t even need two inceptors — Skyryse, for example, plans to certify a single-stick flight control system — but “we did a specific trial and decided this is the right form for the pilot to control the aircraft,” said Wilson.

In a conventional aircraft with hydro-mechanical flight controls, the form and feel of the pilot’s controls are generally dictated by physics. In a modern fly-by-wire aircraft, however, designers essentially have a blank slate — which, if they choose to exploit it, can turn the process of aircraft development on its head. In the case of the F-35, said Lippert, “the preponderance of the F-35 control law, or at least the initial stab at it, was developed before we had an airplane. So, there was a good idea of how we wanted the plane to be able to fly, and that was iterated over time.”

One consequence of this approach is that the development of the F-35 simulator was integral to the creation of the aircraft in a way that simulators hadn’t been in the past. Flight simulators for legacy aircraft attempt to replicate the properties of an aircraft that already exists and handles in a certain way. Some simulators do this convincingly enough that the FAA will certify them as Level C or D and allow pilots to complete their type rating for the aircraft entirely in the sim. At a fundamental level, however, even these advanced simulators are just imitations of the real thing.

There is no full-motion flight simulator today that can convincingly replicate the experience of traveling at Mach 1.6, and thus the F-35 simulator dispenses with motion entirely. Yet in other respects the simulator is extremely realistic, because it uses the same control law that is in the actual aircraft.

“The F-35 simulator flies a lot more like an F-35 than a lot of the other simulators fly like the airplanes they’re supposed to represent,” said Lippert. “Some of that is owed to the fact that they share a lot of the same heritage in development, so you’re flying the same control law in both cases. And so the simulator becomes a very comfortable place to practice and develop the procedures for implementing the automation and using that augmented system.”

The new generation of eVTOLs are generally following the same simulator-first development path, in which the virtual model that describes the aircraft is developed ahead of the physical aircraft itself. “It’s quite remarkable to me that we were flying the simulator well ahead of flying the aircraft — it was not the other way around,” Wilson said of Joby’s approach. “We flew the simulator extensively before we had an aircraft because we knew what the aircraft was going to look like and how we wanted it to be able to fly.”

This has implications for Joby’s future training simulators, he pointed out. Joby announced a partnership with simulator manufacturer CAE in 2022 to develop pilot training devices including a full flight simulator. The eVTOL developer will be relying on CAE’s expertise for things like “the dome and the projection system and the cooling and the safety and the fire suppression and the big ladder you’ve got to climb up to get in,” Wilson said. “But the actual core of the model, the actual software piece of it is entirely with us, and we’ve had that for nearly 10 years now.”

As for how accurately the Joby simulator represents the real aircraft, he said, “it’s not possible for them to be very much different since much of the flight controls software that’s in the aircraft is also in our simulator.”

The instructor stays in the loop

Back in 2006 and 2007, I spent about 12 months and 800 flight hours working as an instructor in Robinson R22 helicopters. Like most primary flight instructors, I have various anecdotes about times when students attempted to kill me. Once I had tasked a pre-solo student with performing hovering maneuvers in a strong wind in order to feel how the helicopter handled with changes in relative wind direction. “OK,” I asked after a few minutes of this practice, “where is the wind coming from now?” The student took their right hand off the cyclic control and pointed.

The reaction to this anecdote depends on the extent to which my audience understands how incredibly sensitive and unstable a helicopter like the R22 is; only my quick action in grabbing the flight controls prevented us from crashing. In an aircraft like the R22 or the Harrier, an essential role of the instructor is exactly this: to stave off catastrophe while the student learns how to maintain positive control of the aircraft.

In a highly augmented aircraft like the F-35B or Joby eVTOL, this role is not as critical. Take your hand off the stick while hovering in either of these, and the aircraft will stay where it is. Moreover, because this automation is integral to the design of the aircraft, it won’t fail in the same way that a conventional automatic flight control system will, suddenly leaving the pilot to contend with an unforgiving manual control scheme.

“The beauty of the F-35 is the degraded state is still heavily augmented, and still nowhere near the workload of flying the Harrier,” said Lippert. According to Wilson, Joby’s aircraft doesn’t even have a degraded mode: it will handle exactly the same way even if two of its three flight computers fail. “The key is to build the functions and features that you want, all those automations in at the very lowest level, so even when you’re down to that one flight computer, they’re still available to you,” he said. “And that way the pilots can be trained more easily because it doesn’t matter what failure you had, your approach procedures and your landing procedures haven’t changed — you still come back and do a vertical landing and you still use TRC.”

The ease of flying the aircraft and the availability of a supremely accurate flight simulator are the primary reasons why the F-35 program didn’t bother to build a two-seat training version of the aircraft. But this does not mean that instructors were made redundant. Take away their direct responsibility for preventing the aircraft from crashing, and instructors still contribute in many ways to developing competent and mission-capable aviators: helping them to refine their techniques, learn correct procedures and build good habits that will keep them safe under stress.

These are roles that instructors perform throughout the entirety of the F-35 training program, even after students begin flying on their own. As Lippert put it, “Just because you’re removing the instructor from the cockpit in terms of execution, you’re not removing the instructor from the loop.”

U.S. Marine Corps pilots who learn how to fly the F-35B follow a highly standardized syllabus that is outlined in an associated training and readiness manual. Flight skills development starts in the simulator, and pilots must demonstrate proficiency in specific maneuvers and procedures before advancing through the syllabus. Erik “Rock” Etz, director of strategy and business development for Lockheed Martin’s F-35 training and logistics team, noted that the F-35 full mission sim that is used for operational training has a wrap-around 360-degree visual system — far more than what I had in Pax River — and the same Gen III helmet mounted display that pilots will have in the actual aircraft, in addition to a fully functional and representative cockpit.

Marine Corps F-35 pilots typically log around 30 hours of training time in the sim before going up in the real thing. For their first five real world flights, an instructor will chase them in a separate aircraft to monitor their performance. From there, the syllabus alternates between simulator sessions and F-35 flights as students learn additional tactics and procedures, such as formation flying and air-to-air refueling.

Students come to the F-35 with different backgrounds and experience levels: some are fresh out of initial flight school, while others are accomplished fighter pilots transitioning to a new model. “In general for the F-35A, B and C, the syllabus is very much matched to the needs of the host nation flying the aircraft, what mission sets do they expect to perform, and then also tailorable based upon the pipeline by which that woman or man starts the F-35 training,” said Etz.

“Someone who has not flown a fighter before will see the full syllabus, whereas someone who’s for example transitioning from a legacy fighter platform, an F-16, an F-22, an F-18, F-15, you name it … that syllabus will be shortened and they’ll have a transition syllabus. But all of it very much front-loaded initially on understanding the aircraft, the way that it operates and flies, and its systems and how they interact with the pilot,” he said.

Given his background, it’s not surprising that Wilson, in his role as director of flight standards and training, is planning a similarly comprehensive training program for Joby pilots.

“[One of the] things that we’re keen on is just to do very thorough training,” he said. “We intend to explain a lot of detail as you go through the training. It’s not going to be some sort of superficial job. This is going to be in-depth training where the pilots have got a deep understanding of the subsystems they’re dealing with, and where things can go wrong and what are the early indications of that and what you can do about it.”

Simple flight controls, sophisticated pilots

The most striking aspect of the F-35 training program is that there’s nothing especially revolutionary about it. It benefits from the fact that the F-35 is easy to fly and its simulator is unusually realistic, but it is rooted in the same fundamentals of aircraft systems knowledge and progressive skills development that have guided military and civilian flight training for decades.

The FAA is not wholly unfamiliar with single-seat aircraft. Consider the Kaman K-Max helicopter, which I learned how to fly at the manufacturer’s headquarters in Connecticut in 2014. This peculiar-looking rotorcraft — which was designed exclusively for lifting heavy external loads — is unusual not only in having a single seat, but also twin intermeshing rotors that give it different handling qualities than a traditional helicopter.

My training course spanned two weeks, including an intense ground school and six hours of dual flight instruction in the HH-43 Huskie, a 1950s-era helicopter with a similar intermeshing rotor system that, as I described it, looked like a breadbox and shook like an out-of-balance washing machine. Then it was off to the K-Max, and my instructor supervised me from the ground as I completed the remainder of the flight training syllabus solo.

Sending appropriately trained and qualified pilots off on their own is a basic feature of civilian flight training that begins with a student pilot’s first solo, when they might have less than 20 hours of total flight time. The first cadre of commercial eVTOL pilots will have much more training and experience than this. In its proposed Special Federal Aviation Regulation (SFAR) for powered-lift pilot certification and operations, the FAA provides some relief for obtaining a powered-lift category rating, but only for pilots who already hold a commercial airplane or helicopter pilot certificate with an instrument rating. Moreover, anyone who serves as pilot in command for a commercial Part 135 operator regardless of aircraft type must have at least 500 hours of time as a pilot, which is more than some military pilots have when they transition to the F-35.

Why then is the FAA so insistent on having dual controls? Some of it can be traced to the agency’s late decision to regulate the certification and operation of winged eVTOLs as powered-lift, a pilot certificate category rating that was added to the Federal Aviation Regulations (FARs) in 1997 in anticipation of a wave of civil tiltrotors that never materialized. Consequently, there are already specific flight training requirements associated with the powered-lift category rating, where flight training is defined as training “received from an authorized instructor in flight in an aircraft.” The FAA also contends that dual controls are necessary for powered-lift type ratings and supervised operating experience.

Related: A united eVTOL industry reckons with the FAA’s shift to powered-lift

Yet, it appears from the FAA’s discussion of its proposed SFAR that the agency is not merely enforcing legal technicalities with its requirement for dual controls, but defending a tradition that stretches back more than 80 years. “The requirement for a dual set of controls for flight training in all aircraft originated in 1938,” the FAA wrote. “It is a foundational safety regulation applicable to airplanes, helicopters, and powered-lift alike that prevents an inexperienced person from being solely responsible for the manipulation of the flight controls.”

In comments submitted on the proposed SFAR in August last year, the Air Line Pilots Association, International also jumped into the mix, declaring that the union is “strongly opposed to allowing only a single set of controls preventing an instructor or second pilot from being able to immediately intervene in the control of the powered-lift aircraft.” According to ALPA, it would be “a significant degradation of safety to allow a pilot in training to operate an aircraft in the NAS [national airspace system] without prior flight instruction in that specific powered-lift aircraft from an instructor having the ability to take control of the powered-lift aircraft” — a position that, if applied to helicopters, would rule out any operation of the K-Max and deem me a menace to the NAS.

A hint as to why the eVTOL industry has gotten pushback for developing aircraft with single pilot stations while the utility helicopter industry has not can be found in the FAA’s reference in the SFAR to simplified vehicle operations (SVO). Defined as “the use of automation coupled with human factors best practices to reduce the quantity of trained skills and knowledge that the pilot or operator of an aircraft must acquire to operate the system at the required level of operational safety,” SVO is a concept that proponents of urban air mobility have sometimes invoked to assuage investors’ concerns about pilot availability and training costs if UAM is a runaway success.

In 2020, McKinsey estimated that the UAM industry could require as many as 60,000 pilots by 2028 (assuming the launch of commercial eVTOL operations in 2023, which turned out to be overly optimistic). By reducing the quantity of trained skills and knowledge required, SVO offered a theoretical path to finding and training these theoretical pilots at a cost that wouldn’t undermine the UAM business model.

SVO wasn’t conceived exclusively for the benefit of UAM — it could ultimately make many types of general aviation more affordable and accessible. In doing so, it could threaten the status and bargaining power of commercial pilots, whose profession today is largely defined by its high barriers to entry. Yet, the vision of SVO that was embraced by investors is also a long way from becoming a reality, as the FAA acknowledged when it wrote, “Some manufacturers are in the process of demonstrating advanced automation technology as part of this movement; however, nothing has been certified yet.”

Where the FAA makes a questionable leap is in assuming that any advanced automation technology employed in an aircraft with a single pilot station necessarily implies that pilots will have reduced levels of skills and knowledge. Simply by virtue of the fact that the FAA expects the first cadre of commercial eVTOL pilots to already hold a commercial airplane or helicopter pilot certificate, it is apparent that they will be required to have more trained skills and knowledge, not less. And even if companies like Joby weren’t already planning on having highly standardized and comprehensive training programs, the FAA is within its power to require them.

According to Lockheed Martin, as of March 2024, more than 2,360 pilots had been trained on the F-35. Since the aircraft began flying almost 20 years ago, the global fleet of all variants has accumulated more than 811,000 flight hours with only one pilot fatality (in an A model) and seven confirmed destroyed aircraft.

Because of the diversity of F-35 models and operators, variations in how accidents are classified and the multiplicity of accident causes, it can be difficult to pin down and compare accident rates. However, the accident record for the F-35 compares favorably to that of other military aircraft with two-seat training versions, including the AV-8B Harrier, and the overall design of the F-35 training program has not been called into question.

The operational environment for the F-35 is very different from the one for an electric air taxi. Nevertheless, the demonstrated effectiveness of the F-35 training program in the military context could help support a finding that eVTOLs with similarly augmented flight controls and a single pilot station can be equivalent in safety to aircraft with dual controls, given an appropriate training syllabus.

Making good pilots better

When I asked Lippert to discuss what he thought accounted for the success of the F-35 pilot training program, he mentioned several factors that I was expecting, including the ease of flying the aircraft and the fidelity of the simulator model. He also mentioned one factor that I didn’t expect — pilot selection.

“In Naval aviation, the process very rapidly determines those who are hungry and who have the real drive,” he said. “One of the biggest differences between military training and civilian flight training is we have a limited number of failures, right? So if you really want to be there, you’ve got to work hard. And you’ve got to show up every day with a good attitude, having done the work.”

In the civilian world, automation is strongly associated with a shift away from traditional piloting skills. Sometimes this is cast in a positive light — as with SVO making flying more accessible and affordable — and sometimes in a negative one, as with the degradation of manual flying skills in commercial airline pilots. But in the world of military aviation, which remains highly selective and competitive, no one talks about the automation in the F-35 enabling anyone to fly it. Instead, the aircraft’s highly augmented flight controls are portrayed as a tool for making good pilots even better.

“There’s [an] overall training consideration in the military, we don’t want to be focusing so much of our time on training pilots to take off and land, we want to be training pilots to go out and perform combat operations,” Lippert said. “So the ability to take that time in training and focus more on the combat piece and less on what we would classically call administrative aspects of flight such as take-off and landing, we want to be able to do that. And having these heavily augmented control systems with layers of automation that support that is tremendously helpful.”

Just as a single-seat aircraft provides the opportunity to think more expansively about the role and duties of the flight instructor, a heavily augmented one prompts the question of what piloting skills remain essential when manual flying skills become less important. These could include things like situational awareness, judgment, decision-making, and the ability to stay calm and respond correctly in an emergency. All of these qualities will be critical for ensuring the safety of high-tempo urban air mobility operations, and all of them can be impeded when a pilot is preoccupied with the “administrative” aspects of flight.

“We do want our pilots not to have to be so in the weeds with limitations and edge-of-the-envelope difficulties … so they are unencumbered by things that can hold you up cognitively,” Wilson said. “If we can make the aircraft super easy to fly, which we have, and if we can make it so that the pilot is highly competent at doing that, so that it’s second nature … then you can lift your pilot’s entire perspective to a higher level and you can introduce more safety because they’re just more prepared for anything that might happen.”

Manual flying ability and the financial resources needed to develop it over many hours in real aircraft have historically been the greatest barriers to a civilian career as a pilot. In this new landscape of highly augmented aircraft and high-fidelity simulators, both factors become less important, and the skills that matter most are likely to be more widely distributed through the general population. The interesting implication of this is that opening pilot careers to many more people — as the eVTOL industry has promised to do — need not result in a degradation of professional standards, so long as training and qualification programs are structured appropriately.

In the Marine Corps, the advent of the F-35 has not fundamentally redefined what it means to be a pilot, and Lippert does not expect that to change anytime soon.

“There is an element of, ‘I want to be good, I’m going to get in the sim, I’m going to practice,’ which I think most aviators have, military or not,” he said, dispensing wisdom equally applicable to a coming generation of eVTOL pilots. “That is a crucial element that will never go away — the pilot is still very critical and their focus, their drive, their desire to perform will always be a top requirement.”

Editor’s Note: This story has been updated to clarify that the RHI on the Joby aircraft commands flight path rate.

Write to Elan Head at [email protected]