This year’s graduating cadre of young cadets could possibly be the last to enjoy full, four-decade careers as traditional commercial airline pilots. By the time they are ready to retire, around 2060, pilot jobs as we currently know them will “start to become obsolete,” according to Richard de Crespigny. He’s the Qantas captain who led a five-pilot team that safely landed a severely crippled A380 in Singapore in 2010.
Pilotless aircraft “will eventually be built,” he predicted, perhaps in production by 2040. “Innovative airlines will buy them. Adventurous passengers will fly them.” Boeing is conducting flight tests related to autonomous or reduced-crew civil aircraft in a two-year project at Moses Lake, Washington. The tests include using a modified vehicle for an autonomous taxi, autonomous flight algorithms in a simulator, flight tests of an artificial intelligence (AI)-based system in a Cessna Caravan, plus engine start, pushback, taxi, manoeuver and takeoff roll using a Boeing 787 Dreamliner technology testbed. The objective is to determine if such aircraft could be operated for freight or passenger-carrying missions with the same levels of safety as current manned aircraft.
“A pilotless airliner is going to come; it’s just a question of when,” said James Albaugh – in 2011 when he was president and CEO of Boeing Commercial Airlines. “You’ll see it in freighters first, over water probably, landing very close to the shore.” Sixteen years ago in 2002, Craig J. Mundie, then chief research and strategy officer for Microsoft (now senior advisor to the CEO), made a $2,000 public wager with Eric Schmidt, then CEO of Google, that “by 2030, commercial passengers will routinely fly in pilotless planes.” Mundie’s logic was that “if we stay on this Moore’s law kick – (computers) will be about 4,000 times more powerful,” adding that “with computers increasingly a part of critical infrastructure, the industry is going to have to focus a lot more on making machines that just don’t fail.” Schmidt’s contrary argument claimed, “the training and timing around handling emergencies such as engine failure at rotation are not going to be transferrable to autopilots and machines.”
“We are quite confident that technologically, the toolkit is filled. With respect to a commercial aircraft, there is no doubt in our minds that we can solve the problem of autonomous flight,” stated John Tracy, Boeing’s chief technology officer, now retired, two years ago. “It’s a question of certification procedures, regulatory requirements and, even more significantly, public perception.”
“The future pilot will still be needed, but he or she will sit in an office flying and managing the aircraft from the ground like the drone pilots already do,” said Capt. Tilmann Gabriel, chairman of the International Pilot Training Association (IPTA), which assists ICAO with expertise in developing training and simulation guidance.
Not surprisingly, many experts believe commercial airliners will never go pilotless. What happens if the computer flying the plane malfunctions? According to NASA data, an aircraft system malfunction occurs on 20% of flights, Moreover, AI would follow “by the book” rules-based judgments and might not be able to make human-type “generative intelligence” decisions such as the emergency ditching in the Hudson River by Capt. Chesley Sullenberger. “Humans are particularly good at adaptive problem-solving and discovery, areas where there has been little machine intelligence progress,” explained Michael Feary, a research scientist at NASA Ames Research Center.
Scenarios of “cascading failures” occur many times, noted Keith Hagy, the Air Line Pilots Association (ALPA) director of engineering and safety. “Those are the kind of abnormal situations when you really need a pilot on board with that judgment and experience and to make decisions.”
A skeptical public
“The issue has never been could you automate an aircraft and fly it autonomously?” said Dr. R. John Hansman, professor of aeronautics and astronautics at the Massachusetts Institute of Technology and Director of the MIT International Center for Air Transportation. “The issue is could you put paying customers in the back of that aircraft”.
A majority of people, 54%, said they would not fly in a pilotless plane (even if the ticket price was cheaper), according to a UBS survey last year. Only 17% said they would, though this percentage increased for those with higher educations and incomes. In the 25-34 age group, 30% responded that they were very likely or somewhat likely to fly in an autonomous airliner (versus 40% who were unlikely or very unlikely to do so). In the 55-64 bracket, the unlikely group rose to 60% with only 10% likely or unlikely.
Americans are far more interested with 27% very/somewhat likely to fly in a pilotless plane, compared with only 15% for respondents from the UK, France and Germany. The 8,000 people surveyed are more inclined to try a driverless car (30%), though the research was conducted before a pedestrian was killed by an Uber test vehicle operating in autonomous mode. In a Travelzoo survey, 38% said they would prefer a Star Trek-style teleporter (at the moment, a fictional device).
The UBS report claimed the aviation industry could save up to $35 billion a year by eliminating pilots in the cockpit: more than $26 billion in pilot salaries, benefits and training costs, $3 billion for business aviation, $2.1 billion for civil helicopters, over $1 billion in fuel savings from computer-driven flight optimisation, plus $3 billion from lower insurance premiums.
How about one pilot?
“The more disruptive approach is to say maybe we can reduce the crew needs for our future aircraft,” Airbus chief technology officer Paul Eremenko said recently. “We’re pursuing single-pilot operation as a potential option and a lot of the technologies needed to make that happen have also put us on the path towards unpiloted operation.”
Boeing Research and Technology vice president Charles Toups said it may take a “couple of decades” to persuade passengers to take a single-pilot jet, suggesting public support for the concept would start with proliferation of self-driving cars.
“They are going to remove the co-pilot,” stated Stephen Rice, a human factors professor at Embry-Riddle Aeronautical University. “The manufacturers like the idea because they can redesign the cockpit. The airlines I’ve spoken to like the idea because it saves costs in the same way that removing the third person from the cockpit did decades ago.”
The future pilot will still be needed, but he or she will sit in an office flying and managing the aircraft from the ground like the drone pilots already do.
To support a single-pilot cockpit, French Air and Space Academy (AAE) and NASA Ames/Rockwell Collins research recommends a ground-based operator, much like today’s military drone operators who control aircraft from half a world away. According to Jean Broquet, an AAE member and former designer of automated satellite control systems, Pilot-Ground operators (PGs), would be qualified as pilots, including holding a type rating (there goes some of those personnel and training cost savings.)
The AAE estimates one PG can simultaneously manage up to five flights in short- to medium-haul operations. In the NASA “super-dispatcher” concept, a trained pilot could remotely oversee the flights of as many as a dozen aircraft at once. If an airborne pilot needed help because of equipment malfunction or medical emergency, the ground-based aviator could help fly the aircraft.
The airline flight crews who participated in their single-pilot simulator-based research “weren’t as negative as I thought they would be,” said NASA research psychologist Walter Johnson. “They don’t want to fly alone, but what I got from them was that, [with a copilot on the ground], it probably would work.”
“The main issue for single-pilot operations is cybersecurity,” said Joel B. Lachter, NASA computer scientist. “In order for it to be done safely, automation or ground operators would need authority to be able to step in in the case of off-nominal issues such as pilot incapacitation. If they can eliminate the cybersecurity threats surrounding those operations, I think it is feasible.”
The chief executive of the International Air Transport Association (IATA), Alexandre de Juniac, said he is “not convinced by the single-pilot issue. I don’t see the plus. I do see the minus.”
Flying above traffic jams
Mashups of technology companies, automotive companies, traditional aviation OEMs and startups are attempting to create a new market niche known as “urban air mobility” (UAM), aka “flying cars” or “flying taxis.”
Google is testing a two-person, electric powered air taxi in New Zealand known as Cora, which uses three onboard computers to calculate its flight path – no pilot necessary – with 12 lift fans for vertical takeoff and horizontal flight, and with a parachute (similar to the Cirrus fixed-wing aircraft). Top speed 93 mph. Altitude 3,000 feet. Range 62 miles. German startup Lilium, which has recruited key personnel from Airbus and Tesla, is touting a five-seat vertical takeoff and landing (VTOL) electric jet which could fly in excess of 180 mph and transit from London to Paris in an hour. The first “functional crewed flight” is expected by next year and the on-demand air taxi service by 2025.
Pilotless aircraft will eventually be built...innovative airlines will buy them. Adventurous passengers will fly them.
Airbus has multiple short-haul aircraft ventures spinning up. Project Vahana from their Silicon Valley-based A3 research centre is a self-piloted single passenger VTOL with automated obstacle detection; a production model is targeted for 2020. The CityAirbus with four ducted fans would seat 3-4, operated by a single pilot (but evolving to fully autonomous operation in the future) – look for a demonstrator and a piloted test flight in 2019. Airbus is also working with Audi on the Pop.Up Next driverless car/drone/air taxi.
A Dutch company unveiled the first production model PAL-V Liberty flying car at the Geneva, Switzerland motor show in March. The $615,000 two-engine, two-seater can drive up to 105 mph on three wheels or fly at 112 mph using a rear-mounted propeller, traversing up to 300 miles on unleaded gasoline. Pal-V requires a short runway to take off and land, and operators will have to qualify as gyroplane pilots (there’s an unpowered rotor which serves as a parachute).
Bell Helicopter (Fort Worth, Texas) is partnered with ride-share company Uber in a flying taxi initiative called Elevate, scheduled to debut in Dallas-Fort Worth, Los Angeles and Dubai in 2020. Initially, traditional pilots will be required, but they are hoping that “pilot augmentation technology will significantly reduce pilot skill requirements, and this could lead to a commensurate reduction in training time,” similar to an FAA light-sport pilot licence. “Not only must the FAA be convinced, but the insurers who cover the risk of the operation will need to see that pilot skill and experience requirements are reduced.”
Lowering the boom
Several commercial supersonic aircraft programmes are on the drawing boards in the US, Japan and Russia, and some of the initiatives are transitioning to a flight demonstration phase. Spike Aerospace (Boston, Massachusetts) flew an unmanned subscale version of their S-512 supersonic design in 2017. Denver, Colorado-based Boom Supersonic’s two-crew, one-third scale XB-1 “Baby Boom” demonstrator is scheduled to fly sometime of this year. Aerion (Reno, Nevada), which has worked with Airbus, General Electric and Lockheed Martin, plans to proceed directly to full-size pre-production AS2 aircraft. All three programmes are targeting flight testing and initial production aircraft deliveries in the 2021-2025 timeframe.
To get there, the contenders not only need to design for Mach 1.2 to 2.2 speeds, they must address challenges of fuel efficiency, carbon dioxide emissions, noise levels on takeoff and landing, and the ultimate dilemma, sonic booms. Even so, they may not be allowed to overfly the US because of an FAA ban on flights exceeding Mach 1 that was implemented in 1973. The FAA is collaborating with ICAO technical working groups on recommended standards for noise and emissions, Mach cut-off (flight conditions in which sonic booms do not reach the ground’s surface) and “low boom.”
Dan Nale, Gulfstream senior vice president of programs, engineering and test, believes flying a supersonic business jet will be “very similar” to flying a subsonic jet. “Modern flight controls and fly-by-wire systems do an excellent job of compensating for the aerodynamic effects of transonic and supersonic flight. Where supersonic pilots are most likely to feel the effects of speed is on their displays. Distances will shrink at a surprising rate, airborne traffic will be quickly overtaken, and pilots will have to carefully plan their arrivals and descents to effectively transition from high and fast to the terminal environment.”
This article has been reprinted and adapted by permission of Civil Aviation Training (CAT) magazine, Halldale Group.
Rick Adams, Editor of the ICAO Journal, is a Fellow of the Royal Aeronautical Society (FRAeS).