Enlarge /. A Cessna Grand Caravan that was flown as photographed by a pilot. But…
The conditions are not ideal for our landing. A harsh wind is blowing over the low hills east of San Francisco and at exactly the wrong angle – right over the runway we're supposed to land on. But as we near our final approach, our two-wing shadow cutting off the suburban homes below makes a gentle suggestion. “I like to do it with my hands. Like a roller coaster, ”he says.
He takes his hands off the controls of our plane, a 27-year-old Cessna caravan that once carried United Nations dignitaries in southern Africa. It's nothing special, with aspects that feel more like a go-kart than an airliner. The cockpit is filled with manual switches and analog dials. Pulleys connect the pedal directly to the rudder at the stern. But recently this aircraft has undergone some modifications. As we descend over 500 feet, the 15-knot gusts hit our side, and the pilot's hands still floating, the wheel and pedals begin to push, compensating for the wind with inhuman precision. The descent remains smooth – even when we touch down.
"It will be very uneventful, almost boring," Maxime Gariel, Xwing's Chief Technology Officer, assured me shortly before our completely autonomous take-off, flight and landing. "This is what we aim for." That didn't seem to mean much about Gariel, an aerospace engineer whose interest in airplanes began with jumping out of them for recreation. But "almost boring" is an accurate assessment. After all, the last thing anyone expects from pilotless air travel is excitement.
Automation is nothing new to air travel. In commercial aircraft, the role of the pilot in handling the aircraft ends largely shortly after take-off. Then the autopilot takes over, as it has for decades. Contrary to popular belief, many modern aircraft are designed to give less control to the pilot in an emergency. Automated flight systems generally cope with changing flight conditions more easily and more safely than humans.
Marc Piette, CEO of Xwing, tells me that these features are far from being pilot-free. On the one hand, there are aspects of the flight that are not yet automated: for example, maneuvering on the taxiways and taking off. In addition, the “autoland” functions common on larger jets typically require ground-based systems to guide the aircraft safely home. The plane cannot do it alone. Even the autopilot still has a big crutch in the traditional sense: the pilot. The challenge is not so much to replace your role in flight, but to replace your role as a communicator. The pilot's main job is to take instructions from air traffic control – to avoid a brewing tower or find a favorable wind, or to dodge the incoming 747 – and adjust the automated system accordingly. It's routine and critical to keeping public airspace safe for everyone.
This role cannot be fully automated. In Xwing's vision, pilots would be replaced by ground-based air traffic controllers, similar to military drone operators who would monitor the flight and set its autopilot in the direction of air traffic control. The goal is to automate as much as possible – the taxi and takeoff, landing and avoiding collisions in between – but keep a human informed. Instead of overseeing one flight a day, pilots could manage many in quick succession – or, who knows, maybe even juggle a few planes at a time. The basic idea: more planes, fewer pilots.
It all sounds pretty reasonable, even straightforward, until you have to plan for something to go wrong. The biggest challenge in removing pilots from the cockpit, says Piette, is a question of contingency: if the operator loses contact, can the aircraft fly safely on its own?
The Cessna is an unusual way to meet this challenge. Much of the focus in autonomous flight is on small drones: quadcopter machines and the like that lack a cockpit or pulleys for human limbs. Piette argues, however, that the humble Cessna, with its mechanical simplicity and long safety record, is a clever way to automate larger, more traditional aircraft. Xwings Cessna has a few extra bells and whistles, but no major renovations. Lidar sensors are attached to the wings with which the strips on the asphalt can be read and guided from the terminal to the runway. There are also visual cameras and radars to spot other aircraft. Machines in the stomach manipulate the flight controls above.
Piette originally had a more streamlined vision of autonomous flight in mind: autonomous planes that take weekend travelers from San Francisco to the pristine redwoods, a six-hour drive north. But he realized that the cargo had more immediate prospects. The caravan is a workhorse of "Cargo Feeder Networks" – regional transport companies that transport packages on behalf of FedEx and UPS in short jumps from larger airports to small cities. They also have a particular problem that Piette believes automation will solve: feeder lines, with their small planes and unusual routes, often face staffing problems. “Nobody wants to fly them. You just have to plan the hours and try to get on the big airliners, ”says Piette.
That's why Piette is busy transforming Xwing into a cargo airline through a licensed subsidiary. In the coming months, some other old Cessnas will be bought and equipped with servers and sensors. Then they will ship things like any other cargo network – only the pilot on board will have very little to do with the automated systems showing the way. In the meantime, they will keep improving their software and using these flights to collect data and prove that their automation systems will work for thousands of flight hours, not just the dozen they have so far. Piette hopes the Federal Aviation Administration will eventually leave the pilots behind him.
That is still very hopeful at this point. The FAA has worked with aerospace companies and academic researchers for years to address the technical and regulatory challenges of unmanned flight. However, with the exception of a few limited licenses and experimentation, it remains difficult to fly autonomous aircraft out of an operator's line of sight. “I am skeptical of the FAA's willingness to tolerate risk in the name of innovation. They say, "We want to incorporate unmanned aerial systems," but that's what they have been saying for at least 10 years, "says Steve Calandrillo, a law professor at the University of Washington who studies drone regulations.
"The challenge is that there is no track record of how secure these systems are," says Cathy Cahill, director of the University of Alaska's Center for Unmanned Aircraft Systems Integration. “The FAA rules and regulations were written in blood. And they no longer want to write in blood. So you are very careful. "
The FAA's first concern is the safety of the flight systems themselves – regardless of whether an autonomous system simply falls from the sky or not. The bigger problem, explains Cahill, is what is called "command and control" – the relationship between a pilot on the ground and the robot in the sky. Autonomous systems that move out of the operator's line of sight depend on a data link between the aircraft and the controls on the ground. In this way, at the request of air traffic control, the air traffic controller can change the flight path and keep an eye on the area around the aircraft using cameras on board. The FAA wants to know how remote operators plan to make this connection so the bird doesn't fly blind. One answer is redundancy. In the Arctic, where Cahill's team sends drones to inspect pipelines and photograph polar seal pups, the planes are connected to the ground through three different channels, including an Iridium satellite and two radio links.
But what if all of these links are cut off? Cahill's team worked with the FAA to validate so-called detection and avoidance systems that identify hazards in the air. These range from acoustics to radar to visual and infrared cameras. The task is easier than, for example, putting self-driving cars on the road with uncomfortable pedestrians and human drivers who break the rules. However, the consequences of failure are worse. She says the technology is close, but not yet proven for large-scale use. Xwing has developed its own system with the aerospace company Bell and with funds from NASA, which is to be demonstrated this fall.
However, according to Cahill, there is incremental progress, approved on a case-by-case basis, that will allow operators to operate out of the operator's line of sight at any given time and location. Last year, the FAA gave UPS and Wing, a subsidiary of Google's parent alphabet, these permits for small drones – primarily used to carry blood and medical supplies. "You used to suggest one of these operations and the answer was 'Hell no'. And then it turned to no. And then maybe it was. And now it has become yes," she says. It's unclear what the FAA made of bigger Will make airplanes like the Cessna, she says, but notes that she could be more comfortable with Heaven's familiar workhorse. She personally would love autonomous Cessnas to deliver packages in rural Alaska, where she lives. The big cargo airline that delivers there went bankrupt last year, and human-operated flights are both expensive and dangerous. "For us, it's an immediate need," she adds.
Piette's vision of a sky buzzing with drones will likely have to wait. "I think the next jump that everyone wants is going to take more time," says Cahill. "I think it will be in the next five to ten years." That's because real infrastructure is required. Think of extensive networks of redundant data connections in the national airspace that are protected from hackers. There will be studies on how pilots should be trained and how many planes they can handle. And, in all likelihood, a much larger public debate about where and how these systems can be used.
In the meantime, people stay on board. As we drive quietly over the San Joaquin-Sacramento Delta, Gariel sits in the back of the aircraft in front of two screens and plays the role of the ground-based "pilot". The detection system picks up some small planes in sight and warns where not to go to avoid disturbing the other planes. But it's a quiet day and there are no immediate threats. In fact, Gariel doesn't have much to do at all. He admits the flights get a little boring at times. But he's hoping for many more boring flights that would prove he wasn't needed up here at all. In the meantime, he thinks, maybe he could start jumping back onto the tarmac.
The article first appeared on wired.com.