Will Future Autonomous Cars Fly Like Birds or Tunnel Like Moles?

It is kind of amazing how much advancement is going on in the autonomous car space. A

• Togluulagch 11 player

year ago, we were mostly talking about cars that seemed comparatively boring, because they just drove on the surface. How quaint -- how 2016. Now when we mention "boring," we may be talking about Elon Musk's new underground tunneling idea. However, a little company called "Airbus" -- yes, the one that likely made the plane you last few in -- disagrees. It plans to start testing its autonomous flying car this year in Silicon Valley -- I expect so it can buzz Musk's Tesla plant there. That kind of explains why Ford just fired its CEO (who wasn't planning to tunnel or fly), and why Toyota is looking at blockchain for automotive security (because you sure as heck don't want flying cars to have Hitchcock Birds). Read on for more about the future of cars, er people-carrying drones, er automated sleds -- ah, personal transportation this week! I'll close with my product of the week: a new drone using Intel's Movidius Myriad 2 technology from DJI, the Spark. (No, it won't carry you to work, but it may become your favorite summer toy.) Flying or Tunneling? This is really both pretty amazing and pretty annoying. I mean, why the heck doesn't someone just invent the Star Trek. Transporter and call it a day? In one corner there's Musk, who has been badmouthing flying cars and just bought an automated tunnel-making machine, arguing that the future is underground. At the same time, a ton more people -- including Google's founders (who are Musk's neighbors) -- a number of startups, and at least one huge aerospace company are arguing for flying cars. Oh, and given that Amazon is developing heavy-lift delivery drones that aren't far removed from this, it's no doubt working on something similar in secret. The Underground Approach So, who is right? Will we soar into the heavens or scurry underground? Well, flying is both easier and more problematic. If we were talking greenfield, I think Musk would be right. The reason is that tunneling is certainly (well, except for earthquakes) safer. His idea of using sleds that would carry cars at high speed is basically a more advanced elevator, and elevators currently are the safest way to travel at speed. You are basically on a rail in a tunnel, so any accidents likely would be either computer- or act-of-god related. There could be a weather impact, resulting from a combination of flooding and pump failure. It happens to subways, but very infrequently. Still, you have to dig the damn tunnels -- and that's expensive and complicated. You'd have to dig them under buildings and under areas where you have power lines, aging waste disposal systems, water pipes, gas pipes and subways. Oh, and there are skyscrapers, which often have support structures that are drilled down into bedrock, and if you break those structures, they will fall over. Even if you could get the permits, which is really iffy, the time it would take to build all of those tunnels likely would be measured in centuries. By the time you were even 20 percent done, you'd likely find the tech you were planning to use and that you designed around had become obsolete. Now, if you were building a new city and could drop the tunnels in first, that would be a different story -- so on Mars, if we ever go there, I'd expect Musk's concept to be top of the list. Here on Earth, though, it would just be too expensive and too slow to get to critical mass. This is largely why Musk stands alone, at least for now. Flying High That is not to say that flying doesn't have issues. Granted, earthquakes wouldn't be a concern. (According to Musk, they wouldn't be in tunnels either, but...) However, weather is a real problem if you are up in the air. Let's say you're in a flying transport with a top speed of 70 miles per hour and a storm hits with 75 mph winds. You'd be pretty screwed, because you wouldn't be able to land, you wouldn't be able to fly out of the storm, and eventually the batteries would run out. (That parachute, instead of saving you, likely would kill you.) That means any hint of a storm like that, and the system would be grounded -- but we don't always get a ton of warning. Musk's tunnel could be hard-wired for power, but flying transports would operate off batteries that would have to be recharged. This technology has been advancing, but it still means not only that they would be severely range-limited, but also that they often would be out of service for charging. An alternative could be expensive battery-swapping stations (with some automated way to move batteries around in order to balance inventory). Ironically, Musk himself has a decent solution for the battery swap, but logistics remain an issue. So, for instance, I live in Bend, Oregon, which is a resort town. Most of the time, we have very little traffic. On a holiday weekend, like Memorial Day, it's hard to get around. So, without some way of moving large battery supplies quickly, we'd have a lot of grounded, battery-less vehicles in town this weekend, because we just wouldn't have the inventory. Finally, if a major problem were to occur in the air, the emergency solution would be to use parachutes or parafoils. With the latter, the vehicle might have a choice of where it landed. Imagine flying over New York, though, and the parachute getting hung up on top of a building, then ripping, then... For planes, which mostly don't fly over tall buildings, this parachute solution makes sense, but over tall buildings we'd need a far better way to deal with a total system failure. (Fortunately, electric power is vastly more reliable than internal combustion, and most designs can lose up to four motors and still remain flying -- but a catastrophic failure, say if the battery died, would be problematic.) Wrapping Up: Why Corning Wins OK I'm kind of messing with your heads calling out Corning, but I'll explain. What it comes down to is this: Technology can and will fix most of the flying issues, and what makes it a favorite to win is that you don't need much physical infrastructure. There are trials going on now, while Musk is still drilling his first test tunnel. We likely can find a way to tunnel faster, but dealing with all the crap we already have underground -- or just the fear that one of these things might topple a skyscraper -- makes his tunneling idea a non-starter outside of a greenfield city where you could tunnel first. Given that we have to solve the same problems for heavy-lift flying drones, the cost of this advancement is spread over far more firms, the problems are being worked by far more companies, and the speed of deployment likely will be far faster. (Dubai is expected to be in limited service next year.) So, flying should win. Now if we build a city on Mars... Hmm, this may explain why he suggested nuking the planet. So, why did I mention Corning? Whether you are talking Musk's sleds or people-carrying drones, you'll need a light transparent material -- either to protect the car on Musk's sled or to see out of in the drone. Right now, the best material for that is Gorilla Glass. This is why it is used in supercars. It is far lighter and far more robust than regular glass, and it doesn't scratch or mar like plastic does, so it wouldn't have to be replaced as often. So, Corning may be the one sure bet in all this mess. Go figure. Rob Enderle's Product of the Week The problem with most drones -- like most cameras, actually, is that you don't have one with you when you want to take a picture of something. The good ones -- and DJI typically is ranked at the top -- simply are too big to lug around all of the time. This brings up portable drones, but they tend to be underpowered and much harder to fly well, and they have really crappy battery life. The one I have, the Dobby Drone, is better than most, but the battery life is still in single digit minutes. Oh, and if it loses its smartphone connection, which mine did last weekend, it crash lands (in my case, breaking one of the propeller blades). The DJI Spark uses Intel's new little AI platform, Movidius Myriad 2, to allow you to control it by hand gestures and automate what it does once it's in the air. DJI Spark DJI Spark It will create 10-second videos automatically by circling you, flying up quickly, and shooting you from overhead while following you. It will carry out an interesting maneuver called "Helix," shooting you as it spirals away. You can keep it flying for a while, as it has a 16-minute flight time, pretty amazing for a small drone. Oh, and it can be launched from, and return to, your hand. The cost is US$499 -- reasonable for an advanced drone. There is a $699 bundle if you want the propeller guards (advisable if you want to do the hand thing), an actual controller, a spare battery, and a bag to put all that crap in. It comes in Alpine White, Sky Blue, Meadow Green, and Lava Red. The DJI Spark starts shipping on June 22 from DJI (and you can order it on Amazon). This DJI Spark is by far -- on paper -- the best small drone currently in market and thus my product of the week (and likely my new favorite summer toy).