This Startup Wants to Fill the Skies With Cargo-Filled Robot Planes
(Bloomberg Businessweek) -- There’s nothing unusual looking about the 38-foot-long cargo plane that’s been flying around Northern California for the past month. But the insides of the Cessna 208 have undergone a sci-fi makeover, resulting in a plane that’s been taxiing, taking off, maneuvering in the air, and landing without a pilot.
The machinery and software that let it fly on its own come from a startup called Reliable Robotics Corp., which has spent four years working on autonomous flight. The company has a grand total of two planes, but its long-term plan is to fill the sky with pilotless aircraft transporting cargo and passengers.
Reliable’s story begins with the self-doubt of its co-founder and Chief Executive Officer Robert Rose. His attempt to become a pilot in college ended for lack of money, but by 2016 he’d earned enough to give the cockpit another shot. Rose, who’d spent his career building autonomous cars and spacecraft for Tesla Inc. and SpaceX, expected that planes would have modernized since he last hopped in a cockpit. But the one he took up had decades-old technology. The shock of how much the flight still relied on a human pilot hit Rose midair as he contemplated his rusty skills and mortality.
“My first thought was, ‘Wow, it’s insane that a private person is allowed to do this,’ ” he says. “You have all this navigation that you need to manage and all the communications you have to do between other planes and taking instructions from air traffic control. There’s layers and layers of stuff. All the while, you are one mistake away from a fatal accident. I kept thinking, ‘How is this OK?’ ”
Rose founded Reliable in 2017 with Juerg Frefel, an old buddy from SpaceX. The pair set up shop in Rose’s garage in Los Altos, Calif., planning to make improved autopilot technology. They hoped to tap into the mechanical and positioning systems available on most planes, buy a couple of off-the-shelf sensors, and tie everything together with clever software that could make the types of decisions usually expected of pilots. Each step of the way, however, they discovered the existing gear for sale wasn’t resilient enough for the job. “You just could not have a serious conversation about removing the human from the plane with these parts,” Rose says. “That meant we had to build.”
These days, Reliable has an office in Mountain View, Calif., where some engineers write software and others craft the electronics, actuators, and other machinery that need to be installed in a plane. The hands-on work takes place at the company’s hangar at the nearby San Martin Airport. During a recent visit, Rose and Frefel revealed the inner workings of their plane to a reporter for the first time, opening up the side of their Cessna to show an aircraft cabin filled with computers and high-precision global positioning systems bolted to the floor. These communicate with custom mechanical devices that control the cables connected to the plane’s elevator, rudder, flaps, and throttle. Because the plane needs backups in case something fails, there are duplicates of almost every part.
It’s typical, of course, for planes to have an autopilot system. Pilots for large passenger jets often perform the takeoff and then let software handle flying and landing. In smaller planes, the pilot can take off, plot a course, and then have the autopilot system manage the adjustments needed to get from point A to point B. The pilots do, however, need to take care of the communications with air traffic control and their compatriots in the air—and step in if something unusual happens. All autopilot systems have been designed with the pilot safeguard in mind; for Rose and Frefel, building a system that reached full autonomy felt more like tackling a whole different problem than making one last incremental step.
Pilot error accounts for more than 70% of fatal accidents, according to federal and commercial data. Reliable’s thesis, which is shared by others in the aviation industry, is that computers can react more quickly and safely than people in an emergency. Pilots are trained to assess a problem and go through a troubleshooting checklist, which sometimes requires them to fumble through a physical manual during flight, finding the right page and following the instructions for whatever midair issue they’ve encountered. Software, which uses sensors and computer chips to figure out the issue, should be able to spring into action and address it immediately.
Reliable’s technology is a long, long way from facing that kind of test. At the moment, a test pilot sits inside its plane to deal with emergencies—and to make the Federal Aviation Administration happy. Communications with air traffic control are handled by a remote pilot at the company’s headquarters. This pilot clears the plane for takeoff with the tower, gives it a route, and then sits back and watches to make sure all goes according to plan. In its test flights in February, Reliable proved for the first time that this long-distance remote operation worked.
The prototype Cessna belongs to FedEx Corp., and Reliable intends to begin flying cargo routes in remote areas first. The idea is that FedEx’s fleet of small planes could be run more often and at a lower cost if the company didn’t need to shuttle pilots around the country and deal with safety regulations limiting how long they can fly. Instead of three pilots flying three round trips in a day, a single remote pilot could oversee the journeys of all the aircraft from behind a computer, Rose says. (The U.S. military already operates drones in a similar fashion.)
Reliable’s systems cost six figures and take weeks to install. The company expects the price to drop as its technology matures and plans to outfit and operate planes for customers and possibly to run its own fleet. It also wants to graduate from remote routes to sending cargo planes above cities, then even to picking up human passengers. People could tap into the U.S.’s vast network of smaller airports and hop about almost as if they had a private plane. “It’s going to take a minimum of 10 years to make this vision realistic, but why can’t regular people just go to the airport, swipe their credit card, get on a plane, and have the thing fly itself?” Rose asks.
He expects that one day Reliable’s mission control will be run by people trained for the job, much like an air traffic controller today, rather than pilots. It’s a less romantic version of flying than the one that draws most people to the job, though some pilots admit there’s a practical reality at play. “An autonomous plane is a great thing,” says Dezso Molnar, an aircraft designer and pilot. “Flying a plane is not difficult, but managing all the rules that can get you thrown in jail is the challenge that most people find daunting.”
Before it can fly autonomous planes at any scale, Reliable needs to prove to the FAA that its technology can deal with all kinds of emergencies through a combination of computer simulations and flights. It also has to get the agency to buy into the use of remote pilots, a concept that has its skeptics. “Remote pilots don’t have contextual understanding when problems occur and can often cause as many problems as they can fix,” says Mark Moore, an aerospace expert who spent decades at NASA and later worked on Uber’s flying taxi technology. “Plus, their survival instincts are not engaged.”
The FAA is under pressure to address not just Reliable’s technology but also an increasing number of new aircraft. Dozens of startups have appeared in the past five years, making electric planes that can take off and land vertically, new types of rockets, and other vehicles. In the robotic realm, Reliable’s competitors include Xwing and Merlin Labs Inc. In a statement, the agency said it was up to the challenge. “The FAA has many initiatives in place to ensure skills of our technical workforce adapt to the ever-changing aerospace system,” said spokeswoman Crystal Essiaw.
For the foreseeable future, Reliable, which has raised more than $30 million, will focus on conducting its autonomous tests and gathering data to present to the government. It also needs to reduce the cost and weight of its equipment. There are other details to sort out, too, such as building computer vision systems that will let the planes drive themselves from the hangar to the runway and back.
Despite these obstacles, Rose hopes to conduct cargo flights by the end of 2022 and considers it inevitable that robotic planes will one day be commonplace. “Pre-Covid, we were moving more stuff and more people around through the air than ever before,” he says. “I think as the costs come down, because of this technology, you are going to see four to five times more flights per day.”
©2021 Bloomberg L.P.