Sikorsky, FedEx Working On Single-Pilot Cargo Plane


Sikorsky and FedEx are reportedly working on technologies to allow single-pilot operations on large commercial aircraft. According to The Air Current (subscription required) the companies are quietly testing the gear on an ATR 42-300 that’s been converted to a freighter. Single-pilot cockpits in commercial aircraft have been discussed for decades but The Air Current says the ATR 42 used in the current tests is the biggest platform to be outfitted with the systems.

The tests are being conducted at Waterbury-Oxford Airport in Connecticut and the publication says they’re part of a broader effort by Sikorsky to develop autonomous systems for helicopters and fixed-wing aircraft. Development of single-pilot airliner class aircraft has been hotly debated in the past 20 years but this appears to be the first practical application of the technology to an airplane of this size.

Russ Niles
Russ Niles is Editor-in-Chief of AVweb. He has been a pilot for 30 years and joined AVweb 22 years ago. He and his wife Marni live in southern British Columbia where they also operate a small winery.

Other AVwebflash Articles


  1. One issue that is rarely mentioned with large airplane single pilot ops is insurance. Who is going to be willing to insure large airplane single pilot operations and at what cost? Does the tightening of the insurance market affect single pilot ops even more? When a company I worked for put a CJ on the certificate they found out that insurance costs were as high as what it cost to pay for a SIC. So the company ended up flying with a SIC on board instead of single pilot.

  2. I think the issue is this: Is AI as safe as having a second officer? To that end, there needs to be data comparing the two situations. Are two pilots as good as one human pilot and one AI system? What about the other way around? One AI system and a human co-pilot or check pilot? Clearly, automation is in the cockpit and the record, so far, is pretty good. Definitely better than it was 30 years ago. But this proposed scenario will need detailed data to cover all contingencies, and then some.

  3. AI isn’t required. What is required is datalink central monitoring. A command center can serve as copilot and observer. If the pilot becomes incapacitated, command center can direct aircraft to land somewhere. Garmin is doing this already with autoland. The technology to maintain the command center datalink and to direct the plane to land exists, that’s not the hard part. The hard part is security of the system against a bad actor either outside the system, or worse, one inside the system (think Germanwings or 9/11 but done remotely).

  4. How does this differ from single pilot authorizations in older light twins or modern VLJs? Reworking the cockpit for one person rather than two seems like evolution rather than revolution?

  5. I’m not an airline pilot so I wouldn’t know, but wouldn’t single pilot ops be a lot more fatiguing? And what about our *ahem* less skilled or disciplined pilots: what happens when they are by themselves for their whole career? Wouldn’t mental health suffer quite a bit, making Germanwings-scenarios more likely and also more feasible?

    • The Germanwings pilot was diagnosed with serious mental problems for years.

      The aviation medical system in Germany was too PC to ground him.

  6. Airplane driver here, and this is the inevitable.. Unlike aircraft development of the past, first introduced as a passenger version and then later as a cargo version.. Single/Zero pilot aircraft will start as a cargo version first to develop technology.. Long before they button up a airplane full of people and send them off on their AI journey..

  7. So what is driving this? Just asking as I am not certain of what might be the underlying issue that is behind single pilot ops. Could it be a pilot shortage? I don’t see the extra cost in hiring another pilot. We all know that pilots don’t exactly become rich flying airplanes. Having said that, just curious. I don’t know the answer, but more automation is coming, no question about that. What is in question, is how fast technology will be implemented, and how it will be applied to air carrier ops.

  8. Electronic systems fail. That’s inevitable. But, the human factor can also fail. Air France 447 proved that with the failed basic skills of two pilots.

    That said, the human factor can be better trained and create higher standards. Do you want to risk the loss of the backup on a popped breaker or one that becomes confused from dual, iced-over pitot tubes?

    Quite frankly, after the arrogance of both Airbus and Boeing having not given pilots and instructors all the information on how a system operates, I don’t trust any manufacturer to insure this is done so pilots are the most prepared possible for any event.

    I’ll always prefer two pilots on board. Never mind the heavier the plane, the faster things happen and the heavier the workload. As for the answer on the poll regarding FedEx has been flying Caravans for years, there’s quite a difference from another thousand pounds over a Skylane with a single turboprop and jumping to as much as eight times that weight with twin turboprops.

    I won’t even get into the impact on the safety of flight with the use of two-crewmember CRM. A second set of eyes along with use of a challenge and response system goes a long ways toward maintaining safety levels.

  9. While pilot salaries are not necessarily excessively high, Social Security contributions, healthcare requirements, unemployment insurance plus injury liability costs, make every employee a drain on corporate profits well beyond the basic salary. This is what is driving the single pilot/no pilot technology. Particularly healthcare. Whatever minimum safety requirements will evolve (just like from the past) by the number of smoking holes, where they are located, and whom they hurt while digging the impact crater.

    My hat is off to all the relatively new pilots flying single pilot IFR in complex twins, single and multi-engine turbines, VLG’s flying multiple short routes, sometimes into or out of high traffic areas or to non-tower airports, including some with basic/minimal IFR capability, with lots of landings and take-offs, usually at night in virtually any kind of weather at gross weight. Seems a bit paradoxical that the newbie gets thrown into the toughest, least forgiving flying environment, while experience with seniority allows for long flights with only a single take off and landing usually into well equipped airports. Ah, the benefits after completing the aviation “rights of passage”.

    I can see how complicated things for a single pilot could get losing an engine on an ATR-42-300, IFR at night, in weather after several hours of up and down flights. Having another set of eyes, hands, feet to identify the problem, engine, and make quick corrections could make this a non-event rather than a grim statistic.

    I am sure someone will pipe in to say AI can do it better and faster. That would be true if software programmers have the ability’s to anticipate and program all potential accident chains. We saw how that worked out with MCAS.

    Somehow, so many think AI or some sort of robotic technology can solve all problems associated in a fluid, 3D environment. Maybe in the ground bound 2D world. However, I don’t think the public who remains underneath all of this aviation autonomy will be comfortable.

    It has been said many times before, insurance will play a key role along with public perception. I agree. Most people do not like flying nor airplanes in general. Drones are proving to be nuisance to many people. But their size determines the threat factor. But when we decide its safe to fly relatively heavy airplanes either single pilot or autonomously, with the public actually knowing the least experienced aviators are behind the controls in a single pilot aircraft, there will be backlash many aviation pundits have potentially not foreseen.

    Right now, the public is clueless that the Baron overhead at night in whatever weather maybe happening locally, is being handled by more often than not be somebody with low time, yet in a highly task saturated flying environment, in an airplane with marginal single engine performance…paying their aviation dues. Make that an ATR-42-300 or any small to medium size bizjet, flown single pilot or autonomously, or should the Baron crash into the local neighborhood, I am sure many will have something unpleasant to say.

    No matter the size of an airplane, the public perceives airplanes are flown by a wealthy person of privilege or by a highly paid, impeccably trained commercial rated pilot. Both imply lots of experience, lots of bells and whistles on board, and all weather capability. J-3’s to 777 all count the same. Share holders being led by bean-counters are driving this horse.