Silicon Valley discovers aviation – but for how long?

Kitty Hawk test flight
Kitty Hawk released a video that shows its flying jet ski in action.

You have to pay close attention these days to keep up with all the breathless news about “flying cars” and “disruptive aerial vehicles.” The great and the good from the technology world have fallen in love with aviation lately, and their various startup companies have been launching aviation projects at an unprecedented rate in 2017. More than modifications to existing airplanes, each new design seems to be more fantastical than the last.

Here are just a few stories that have made news in the last month:

  • Google co-founder Larry Page is behind Kitty Hawk, a company that took the wraps off its “flying car” this week with a sexy video. It looks a whole lot more like a flying jet ski than a car, and it’s actually a Part 103 ultralight so it won’t require a pilot certificate.
  • Not to be outdone, Google’s other co-founder, Sergey Brin, seems to be building a large airship at the NASA Ames Research Center in California. Plans for commercial introduction are still unclear at this point.
  • Skype founder Niklas Zennström also showed off his big idea this month. Lilium, an all-electric flying car with vertical takeoff and landing capabilities, made its first flight in Europe recently and promises some impressive performance numbers.
  • At the swanky Top Marques Monaco event this week, Slovakia-based AeroMobil opened the order book for its supercar-with-wings, which is the closest thing to a real flying car yet. It’s priced at $1.5 million, and requires a pilot – for now.
  • Proving that flashy headlines aren’t just for startups, Boeing and JetBlue announced an investment in ZunumAero, a Seattle company that thinks it can build an electric, short-haul airliner that will cut airfares by 50%.

What precipitated this mad dash into concept aircraft? There are as many theories as there are startups right now, but the most likely ones are also the most mundane. A lot of the gold rush mentality is just the way the venture capital-fueled technology industry works these days. Everyone is paranoid about missing the next Facebook and being left behind, so interest and money quickly shift from trend to trend at the first sight of change. Since one of the hottest trends in recent years has been drones, many of these flying projects can be viewed as basically grown-up versions of quadcopters. Plus, for billionaire founders and dominant companies, a $100 million bet on a crazy aviation startup is about like you and me dropping $10 on a lottery ticket.

Uber concept for VTOL
Uber’s push for airborne taxis by 2020 is driving a lot of the headlines right now.

But perhaps the most important driver of flying car startups has been Uber. The fast-growing, money-losing company has made no secret of its plans to take its taxi service airborne, forever changing urban commutes with a vast fleet of unmanned vertical takeoff and landing (VTOL) aircraft. This effort seems to have brought a lot of hobby projects into the daylight, and a conference hosted by Uber this week shows just how pervasive the drone fever is right now. A diverse group of companies took the stage to announce radical designs and make increasingly bold predictions, including Embraer, Bell Helicopter, and a Mooney-Carter partnership.

Such attention is downright exciting for pilots, who have long made do with minor upgrades to 60-year old airplanes. Finally, the best minds in the world are talking about truly new aircraft and bold ideas! While the buzz certainly is intoxicating, it’s almost guaranteed that this bubble – and it is a bubble – will burst. At least three forces are acting against these aviation startups: regulation, technology, and attention span.

The most obvious obstacle for these aviation startups is the regulatory environment. Tech firms use catchy phrases like, “move fast and break things” to signify their casual disregard for the rules, and that’s often successful (as Uber has rather spectacularly proven). As these companies will soon find out, though, aviation is quite different. Unlike many areas of the technology industry, the rules are well-established – perhaps too well-established – so there are few gray areas to hide in.

One of those rules involves powerplants. Almost all of the recently-announced aircraft use electric motors, which are nearly impossible to certify right now. Strapping eight of them to the wings and then tilting them for takeoff can work, but don’t expect a type certificate for such a design to appear overnight.

On top of the electric motors, the business models for these airborne Ubers depend on autonomous flight if they plan to make money. While the FAA has released final rules for small unmanned aircraft, these only apply to aircraft under 55 lbs. There is a long way to go before a comparable rulebook is in place for 5,000 lb. drones that carry passengers. One key issue to solve is “sense-and-avoid,” the ability for an unmanned aircraft to avoid a mid-air collision with an airplane. There are various ideas for tackling this issue, but formal trials won’t be complete until at least 2020, with the final decision probably even later. For perspective, consider that Europe allowed commercial operators to use single engine turboprops on IFR flights last year, and only after decades of debate and study. Why will autonomous Part 135 or 121 operators move so much faster?

Carter Mooney design
An all-electric, unmanned Slowed Rotor Compound aircraft – should be easy to certify, right?

Perhaps most importantly, unmanned air taxis will have to do battle with a single, nationwide regulator for aviation. While Uber is no stranger to legal battles, until now they have benefited from the fractured nature of their opponents – San Francisco Board of Supervisors here, the state of New York there. When it comes to flying cars, there can be no jurisdictional arbitrage; if the FAA doesn’t approve, it’s not going to happen.

Beyond the legal challenges, there are big technical problems to solve, too. Electric motors, lightweight batteries, tilt rotors, and autonomous flight controllers may not be science fiction at this point, but they’re hardly mature technologies. Many of them depend on continuing development in the smartphone and electric car industries. While the critical components will eventually get made, the key questions are: at what price and on what timeline? In particular, the schedule for these new aircraft is incredibly aggressive. Uber plans to launch its Elevate service by 2020 – a mere 32 months from now. That is nearly unthinkable given the current state of technology, regulation, and public perception.

It’s also worth considering how well these aircraft will solve the problems they hope to take on, notably traffic congestion. As Zennström said recently, “The way we deal with transportation today is broken. There are congestions and to get from East London to West London takes forever. There is pollution in our cities with carbon dioxide so we get climate change.”

VTOL aircraft certainly do eliminate the restrictions imposed by road infrastructure, but they hardly eliminate the idea of traffic. Crucially, Uber isn’t so much solving the capacity issue as it is moving where the problem exists. Take a commonly-used example: New York on a Friday afternoon, when thousands of people want to get from Manhattan to JFK Airport at 5pm. Right now the only real option is an hourlong car trip through heavy traffic, but the cheerleaders suggest that could be cut down to five minutes in an Uber Elevate. In flying time it certainly could, but where will all of these aircraft land? At 4-6 people per Uber, it would take hundreds of drones to satisfy demand. They may be able to fly their own routes, but they will eventually need a place to land. As we’ve learned in aviation, the real bottleneck is often the number of runways, not the airspace.

Uber is working with real estate developers to address this issue, but I suspect they haven’t considered how vocal the NIMBY crowd can be when aircraft are involved. Cars are a part of everyday life; tilt rotor aircraft are not. Even if the neighbors do go along, this will be a very capital-intensive effort. Uber (or its partners) will have to own a lot of land for vertiports, and have a large fleet of aircraft. Neither of those exist in the company’s current business model, which is the ultimate “asset light” strategy. Indeed, one of the key breakthroughs for the company now is making use of existing capacity (garages and cars) that is sits idle most of the day. Building a network of facilities and aircraft from scratch is a different story altogether.

Larry Page
Will investors like Larry Page simply lose interest?

Regulation and technology are perpetual enemies of aircraft development programs, but the third reason for skepticism is less common in this industry: investors will simply lose interest and move onto the next idea. For anyone used to rapid development cycles in software, aviation must feel utterly unbearable with its lengthy testing programs and mind-numbing certification rituals. In fact, the attitude it takes to run a successful technology company is almost the exact opposite of the one it takes to bring a new aircraft to market. It’s easy to see how, after a few years of being stuck in the mud and wasting a few hundred million dollars, billionaire investors will simply punt. That’s not necessarily a criticism, it’s just the way their business works. Technology investors know they will strike out far more often than they will get a hit, but it only takes a few really big hits to win. When it’s clear that a flying car isn’t going to be the next multi-billion dollar company, it will disappear.

None of this is to suggest that autonomous flying cars won’t exist someday. I look forward to riding in one before I die, but the outrageously ambitious schedule being promoted right now is a publicity stunt, not a realistic plan. I suspect most of the people involved know this, but keep up the charade to fuel the hype cycle.

It’s not all doom and gloom. When this bubble pops, there could be some very real benefits that trickle down to general aviation pilots. It seems increasingly likely that the future of recreational aviation, if perhaps not cross-country flying, will be electric. If general aviation manufacturers can free-ride on some of the advances in electric flight and autonomous technologies, we could end up with a new generation of bugsmashers. That’s something to look forward to, no matter what you fly today.

So I welcome our new visitors to the bizarre world of aviation, but I recommend we all keep our hands on our wallets and our eyes on the FAA. When it’s over, maybe we’ll have something to celebrate.

24 Comments

  • The day is approaching rapidly, when autonomous vehicles will render obsolete both drivers’ licenses and pilot certificates. THAT – not electric propulsion – is the disruptive shift-of-paradigm that is going to democratize personal transportation everywhere.

    If the FAA can’t or won’t get with the program, then America will find itself saddled with a third-world transportation infrastructure, as the rest of the world’s regulatory agencies embrace inevitable progress.

    On the electric propulsion front, batteries are fine for toys. But real aircraft – with practical range and payload capabilities – will need another means of producing their electricity. Fortunately, two mature technologies exist: engine-driven generators, and fuel cells.

    All of the pieces are in place. Get ready for an incoming asteriod.

    • I agree that this is the future and I’m excited by it. But do you really think by 2020?

      The FAA seems to be making a serious effort to move faster but recent evidence suggests they are still pretty deliberate. Left behind? It’s already happening with small UAS.

  • The Dubai project says 2020. I’d keep an eye on that one. But Uber? Not a chance.

    In just about every industry I’ve encountered, mixing manufacturing with operations has been a disaster. Wisely, Boeing doesn’t operate an airline, and Southwest doesn’t build airliners.

    Uber thinks that its moment in the sun will come when it no longer relies upon owner-operators. Quite the opposite is likely to happen. But somebody will make the on-demand GA paradigm work on a truly grand scale.

    Uber’s game is short hops. But “UberAir” won’t make much of an impact by trying to fly Gotham denizens to LGA. The world doesn’t want or need a faster ride to the airport. The real impact will occur when / if Uber figures out that the “killer app” for autonomous on-demand GA will be to REPLACE the airliner ride altogether – for trips of 500 miles or less.

    Remember, Edison’s objective was NOT to improve the candle…..

  • Maybe not by 2020 but this will happen and soon. I don’t think the FAA will stand in the way. Silicon Valley has a way of busting through regulations either by ignoring them or paying for them. Things are changing too fast and they won’t be stopped by regulation. They’ll lawyer up and act first, beg forgiveness later. By then the public will see the benefit and the FAA will have to change to suit.

    I do think pilots and others in the industry should buckle up and get ready for an interesting ride over the next 5-10 years. The biggest one I anticipate is sophisticated autopilots (AI) in airliners, eliminating the “pilot shortage” by dropping to one pilot. Replacing helicopter pilots and other single-pilot non-passenger operations won’t be far behind. On the bright side, small electric, affordable private planes will hopefully come next but I doubt they will be “hand flown” in the way we are used to.

  • I expect to see electric aircraft a la Tesla within the next 10 years. Flying cars have been in comic books for nearly a century, so one can expect some iteration will become a reality. It took that long for picture phone to become a reality. The current trial baloons are test objects. I expect a high level of automation, and the flying cars may not even have any controls to fuss with.

    No one can stop the march of progress, not even FAA. As Silicon Valley gets in the game of making campaign contributions, I expect the regulatory regime to change accordingly.

    He who pays the piper calls the tune. Same is true for politicians.

    • You’re right – no one can stop the march. And let me be clear that I can’t wait. I love technology and think most of this stuff will be a huge positive for the world.

      But while the direction is obvious, the timing is not. The FAA can’t stop progress, but it sure can slow it down. So can NIMBYs who are scared of anything that flies.

  • John, your analysis is spot on, but perhaps too generous. There are layers upon layers of challenges– some not yet imagined– to be overcome by the dreamer-advocates of filling the air with flying, autonomous taxis. These barriers will slow their progress and ashcan many programs. As FDA people often say, the FARs are written in blood. It’s hard to see the agency dramatically loosening the aircraft certification and operational rules to accomodate radical ideas about busy low altitude flight over congested, heavily built-up areas. I am reminded of the ancient joke: Q– do you know how to make a small fortune in aviation? A– start with a large fortune.

  • I suspect that Silicon Valley engineers are more accustomed to dealing with average risk; I know this is the case with investors. They may not quite grasp the level of risk aversion present in public transportation by air, and the effect that certain specific risk focus points have on system safety assessment. Take a look at Figure 2 in Advisory Circular 23.1309-1E, which is titled “Relationship Among Airplane Classes, Probabilities, Severity Of Failure Conditions, And Software And Complex Hardware DAL”. Before we predict that the regulatory environment is going to change, we should consider how these numbers came to be, particularly those in the right hand column describing the relationships between the terms “catastrophic” and “extremely improbable”. These relationships represent a fundamental standard of public risk; why would that change? If it does not change, what actually has to be done with a new technology to meet those requirements?

    Next, consider what happens to those probabilities when the live pilot is removed from the equation. One of the most interesting aspects of the autonomous discussion is that while we have reams of data about what pilots do wrong, and can often write code to prevent those specific errors, we have very little data about what pilots do right. We really have no way to measure and catalog the thousands of tiny decisions that a pilot makes on each flight, and what effect those decisions may have on a safe outcome. All we know is that the outcome of a particular flight was safe. In order to remove the live pilot, or even transition from a two-pilot cockpit to a single pilot, we will have to account for the absence of a human system that we know makes those thousands of tiny decisions on each flight, about which we statistically know next to nothing. Yet this will have to be done with respect to the standards outlined in Figure 2.

  • If you want a good airplane, then get an airplane. If you want a good car, then get a car. If you want a crappy airplane, and you want a crappy car, then a flying car is just the ticket for you!

  • Whilst I am in general agreement with the sentiments expressed here, and whilst I think that fully autonomous ground and aerial vehicles will be with us probably sooner rather than later, I would still like to be able drive and fly as I do now, I enjoy it.

    In the case of recreational aviation, I would still like to be able to manipulate the controls as I do now, perhaps with some fly-by-wire technology as a safety back up for an extreme event, but otherwise let me fly it myself. I do not want to simply climb in and push a few buttons and let it ‘do its thing’. Indeed, if that’s the way forward, then GA and recreational aviation will wither and die. Either that or we will only be allowed to fly simulators. Couple that with the predictions that AI and Expert Systems will make most of us redundant, and perhaps this debate will be a moot point, we will all be unemployed and unemployable, and will simple eke out a subsistence existence. Welcome to the Brave New World – I’d rather not!

    • “I would still like to be able drive and fly as I do now…” The existence of autonomous vehicles won’t stop that from happening, Adrian. AVs WILL allow many more people to participate in GA, including pilots who have lost their flight privileges owing to medical conditions.

  • While flying cars sound great, and I’m sure they’ll be here sooner rather than later, my big concern is using your airplane in everyday traffic on the road.
    Almost every car has numerous dings and dents. If someone bumps into your car at your favorite restaurant and causes minor damage, you exchange documents and continue your vacation. Or have the car towed and get and rent another one. But if you’re driving an airplane, will you need to find an A&P to inspect your vehicle before you fly it again? And if there’s minor damage to the aircraft that precludes flying it, now you leave your airplane maybe thousands of miles away until it’s repaired.
    I think I’ll continue to use an airplane in the air and rent a car when I get on the ground.
    The Kitty Hawk jet ski and other purely recreational vehicles sound neat, but as a reliable transportation option, I’m not so sure.

    • Most “flying cars” will never see any duty as ground transportation vehicles. The reasons you mention are among many.

  • I don’t think flying cars, taxis or whatever will ever really take off. Most people are not afraid of flying in airliners but I can’t see people taking to the skies en mass in small flying vehicles. What about turbulence? Bird strikes? Is the average person equipped to just ignore that stuff while the computer “handles it”? My wife would freak out. My office mate would freak out. I guess my point is everyone’s focusing on the technological aspect of flying cars and ignoring the human aspect. Are they trying to solve a problem that doesn’t really exist? I think most people would think it’s cool that we now finally have flying cars, but most people would also not want to utilize them.

    • “Is the average person equipped to just ignore that stuff while the computer ‘handles it’ ?”
      Well, Scott, is the average airline passenger equipped to ignore the stuff you cited, while the pilots “handle it?” I’d say that most people have that particular “right stuff.”

      “Most people” may never utilize low-occupant-count autonomous aerial vehicles. But then again, “most people” don’t utilize airline travel, either (fly somewhere at least once a year). Nonetheless, commercial airlines are a thriving industry.

      My guess is that most people don’t enjoy a long ride to an airport, where they endure outrageous parking fees, long lines, TSA indignities, the risk of being “bumped,” 30-minute boarding intervals, 20-minute taxi intervals, and a repeat performance at their destination airport. They do it because they don’t have an affordable alternative.

      Point-to-point in a 4-PAX vehicle, sans the above-cited hassles, for trips of 500 miles or less? Not everybody will do it, but enough will.

      BTW, in the early 1800s, most people “freaked out” at the thought of traveling at 20 miles per hour on a passenger train. Some doctors even predicted that the human body could not survive such madness. Thankfully, no one was forced to get aboard at gunpoint! And elevators in high-rise buildings???

  • Has anybody considered the road hazard risk to flying cars? Have a fender bender in the family sedan and it is fixed within several days at local shop at an (insured) reasonable cost. Try that with a flying machine under FAA regulations! A $ 3-5K auto repair might raise to $25-75K in cost and take weeks or months to repair along with the loss of use.

  • Flying cars, hah! Noisy and dangerous machines, perhaps. With a decade or two of refinement I may go so far as to watch someone else fly one!
    The “flying car” thing is far from recent, however. I’ve been following Paul Moller’s quest for many years (and Peter Garrison wrote a not-so-positive article in Flying, perhaps two decades ago, and it appears that the “SkyCar” is no closer to reality today.) Nine figures of US dollars burned, a loud and unstable flying machine–that has never flown untethered–to show for it. Not to mention performance claims that cannot happen (better specific fuel consumption using Wankel engines? Not if they are still made of metal.) Many promise, no deliveries, as I figure most of these other things will become. (For giggles you can go to moller dot com and take a look at things you cannot buy. Heck, you can even invest.)

  • Try building a conventional electric powered light aircraft we can actually afford! That cost should not exceed $50K.

  • I think Terry has it right – keyword “Affordable.” That has just about gone out the window with GA, and these Silicon Valley aces don’t even think of the consequences, until after the fact, when their latest and greatest toy is being vandalized by “gee whiz hackers” proving that technology is not so great after all. At the same time, the FAA will have a lot more to say about this than any genius with a bottomless wallet, and the techno geeks don’t have enough money to buy that sink hole.

    • Ray I agree with you totally. I think there is something missing with the World – it is called ‘common sense’. You can’t solve the problem with congestion with putting more vehicles in service, airborne or not. The only way to solve the problem is to reduce the number of cars on the road AND have an efficient underground metro system. There already exists a subway to take someone from downtown Manhattan to JFK, in 1hr 30mn or so. If this is not fast enough for these people, they can think about reducing the journey time. Whatever it is, when one gets to the congested airport, he will stuck in the security system anyway or be interrupted by John McLean, who will be saving the airport in his Die Hard Part 2 at JFK airport!

  • It’s hard to argue against the likelyhood that automation has improved safety in aviation overall but the overhyped spin about autonomous vehicles (ground and air) being commonplace within 10 years deserves healthy scepticism. Disruptive technology such as the internet and smartphones are used as examples of sweeping change, the difference though was that your 1st generation smartphone frying didn’t kill you.

    See this article for an interesting example of automation going ‘psycho’. It seems that the crew never obtained a satisfying answer as to what actually happened…

    http://www.smh.com.au/good-weekend/the-untold-story-of-qf72-what-happens-when-psycho-automation-leaves-pilots-powerless-20170510-gw26ae.html

  • The problem is energy density. Electric batteries of a given volume cannot store anywhere near as much energy as the same volume of av-fuel (kerosene or gasoline). Also, combustible fuels use air from the atmosphere as part of the total energy, but that air takes up no space and no weight inside the aircraft (a bonus that most people don’t think about, except engineers who design rockets where the oxygen must be carried on board). A battery has to contain the entire chemistry of the energy system. The combustible system only carries the hydrocarbon component and uses air outside the system to complete the chemistry.

    Electric motors are very efficient compared with piston or turbine power plants, but not enough to overcome the fuel (battery) density issue. And as batteries become more dense, they become more dangerous. Li-ion batteries are essentially sticks of dynamite. Any short in the battery can be catastrophic which is why the FAA has restrictions on air shipping them. Just think about all those hoverboard fires and you get the picture of just how dangerous electric batteries can be.

    The key to this problem is the fuel cell. The chemistry of the fuel cell essentially burns the hydrocarbon (or hydrogen) with external air, so the energy density is much better than a battery and similar to that of a turbine or piston engine type system. Since the fuel cell produces electricity, this can be used to run the much more efficient electric motor to drive the propeller.

    If Silicon Valley wants a breakthrough, they need to focus on the fuel cell. That is the key to making electric airplanes practical.

    Jack

  • I am afraid that, as John suggests in his article, the excitement about “uber-air” and other short hop pilotless vehicles (i.e. “the flying car”) is more about falling in love with the technology (or promised technology) than solving the underlying problem of rapidly getting from one random location to another in a densely populated urban area. The fact is that existing streets and arterials, properly upgraded with guideways and means to deliver electric power to the “smart” driverless vehicles that use them, coupled with traditional mass transit systems like light rail, could easily manage considerably more passengers than they currently do, and that gain would be many times the number that could practically be transported in “flying cars.” What’s more, while the technology to enable this integrated terrestrial transportation system would certainly push the envelope it would be vastly less daunting than what would be needed to get massive numbers of flying cars to play nice with their environment and each other.

Leave a Reply

Your email address will not be published. Required fields are marked *