Image: EHang

As has been reported by various media outlets this week (e.g., Fortune, The Telegraph, Engadget), Dubai has announced that it will begin single-passenger autonomous drone taxi operations this summer. The ability to fly to your destination across town, as opposed to driving, is especially appealing in cities that are hyper-congested with road traffic. I visit Los Angeles quite frequently and still remember making the mistake, back in 2008, of getting on the 405 at rush hour and spending two hours to travel a mere 10 miles. And according to an article published in Forbes last year, Los Angeles is only number 10 in the world’s most traffic congested cities.

Although we have had reliable helicopter transportation for well over a half century, they are big, loud and expensive. An electric-powered drone such as the EHang 184 currently being tested in Dubai, could operate in smaller take-off/landing spaces and may potentially be quieter and cheaper.

There are significant drawbacks, though. In an IEEE Spectrum online article yesterday, Evan Ackerman has noted that an autonomous drone taxi could potentially be very dangerous, more so than helicopters or airplanes. When an engine quits, an airplane can glide and a helicopter can autorotate to a safe landing (in the hands of a qualified pilot). EHang claims their drone has full redundancy and it appears they are claiming to have two independent power systems on board for backup purposes. Even assuming this backup system indeed performs flawlessly, 113 years of experience in powered flight have taught us that unanticipated things can, and do, go wrong. A significant amount of pilot training is aimed at how to handle emergencies. Aircraft and automobile manufacturers all have to meet voluminous safety standards in their designs. It is safe to say that the safety standards for autonomous drone taxis have not even been written yet.

Am I a Neo-Luddite? Not at all. It is quite clear that long-haul trucking will cease to be a form of employment long before the end of this century as autonomous vehicles will take over. Ocean-going container ships will similarly become unmanned. However, in these cases, the new technology is effectively a robot driver, not a brand new, relatively untested vehicle. It is worth noting that the first scheduled fixed-wing passenger airline service did not occur until 10 years after the airplane had been invented.

The bottom line, to me anyway, is that the Dubai drone taxi plan is premature. The octocopter design employed by EHang in their 184 works quite well for carrying movie cameras but the potential pitfalls as a passenger-carrier have yet to be explored.