Rather Good GuidesĀ®

Documentation the way it ought to be.

UAVs, the FAA, and the Amaz-o-Copter

 

 

Well, here we are at the end of 2013, and the FAA still has not given much indication of how it is actually going to integrate commercial unmanned airborne systems into the national airspace (NAS). In summary, the current "roadmap" (a curious name for the FAA to use) tells us that UAV pilots, aircraft, and airspace will be regulated. Oh. Right. Well, that's that then. I was rather hoping that (a) it would be an airmap, and (b) we might get a hint, a faint whiff, a tiny scintilla even, of what the regulations might sorta, kinda be?

 

Amazon's Jeff Bezos stimulated a bunch of press reports and discussions when he showed off the concept of having Amazon deliver packages to households using octocopters.

 

Nice ideas. Completely impractical with today's technology. And next year's technology. I think we're going to need a few years, perhaps a decade before we can make them feasible.

 

But right now. Sorry. The ideas in the Amazon video are so bad they're not even wrong.

 

Let me count the ways. Copters have lovely slice'n'dice blades on them. They spin at about 4,000 RPM and slice through human flesh very, very well.  No risks there, right? Also, homes are surrounded by trees, bushes, vegetation, poles, wires, antennae, pets, people, and vehicles -- and quite a few of the objects on that list are prone to move around in arbitrary ways. How would the copter (as they are more properly called) managed to avoid the slicey-dicey routine on you, your kids, your rose bushes, or just the phone line between your house and the street?

 

How could an autonomous copter navigate from an Amazon distribution center to a house just down the street? It couldn't. Yet.

 

What would it need to be able to do it, though? Oh, let's start with LIDAR, a laser-based sensing technology currently being developed by Google for self-driving cars. But such devices are expensive (meh, around $10K) and heavy (both in terms of weight and the electrical requirements). Heavy means more battery power. More battery power means more heavy. Soon you need a second copter to carry the batteries for the first copter. There's an uneasy trend you can see forming here.

 

The UAV will, of course, use GPS to find it's location, but civilian GPS is not that accurate -- typically no worse than 7.8 meters or 25 feet. But given that the smallest twig on a tree can cause a copter to crash, that just isn't good enough. That's especially true on those days when there are solar storms -- yeah, the sun wreaks havoc with the GPS. Solar weather and terrestial weather both play huge factors how well the GPS systems works.

 

OK, so we'll need to beef up the GPS too.  No complaints from anyone on that score. Just hope the terrorists don't have copters when we make the GPS accurate to 20 cms or so.

 

Anyway you slice it (just like the copter's prop blades) we're left with the dilemma that, for the foreseeable future, the requisite technology will be too heavy and too power hungry to actually allow Amaz-O-Copters to get off the ground.

 

How's about making it non-autonomous with a ground-based pilot? Nope. That's not going to work -- the pilot would need to be able to see where the copter was going to fly it and avoid the aforementioned hazards to navigation -- and that would require a so-called FPV (first person video) system. We use them on copters right now -- but they're strictly line-of-sight -- and often not even with sufficient range to do that. FPV cannot be used when the copter is out of line-of-sight -- and the FAA would have to permit long-range FPV systems as well and they're not even close to thinking of thinking about that!

 

So 2013 closes with the commercial UAV world in the USA in a bit of a mess. We don't have regulations that permit anything more than hobby and recreational use. We can't fly copters commercially. We can't fly them above 400 feet above ground. We cannot fly them beyond visual line of sight. And we sure as hell don't have copters that can come faintly close to doing what the Amaz-0-Copter purports to be able to do.

 

And even if the Amazo-o-Copter becomes a reality, I suspect the image below will depict some part of that reality (props to @QuantumPirate on Twitter for creating this):

 

We can but dream. Wealthy people can dream bigger and make things happen because often the problems that must be solved are dollar problems -- throw enough dollars at the problems and they can be solved. And Jeff Bezos has enough dollars and look at how he solved the problems that Amazon.com needed to solve in the early days.

 

But let's hope the drones don't reach sentience as the image above suggests. Read Daniel Suarez's Kill Decision if you want to lie awake a lot and worry about how that whole sentience thing might play out. Hint: It doesn't end well. In fact it doesn't even start well, either. Adds new meaning to the Blue Screen of Death, as you'll find out when you read it -- one of the best high-tech thrillers in recent years. Suarez gets his technology spot on!

 

But thanks, Jeff. You made a lot of people aware of the possibilities. You may even have partly rehabilitated the use of the word drone to mean something friendly rather than a killing or surveillance machine in a war zone. Your ideas are valid, though, but the present implementation sucks. Please do throw your dollars at the problems. I'd be first in line to use Amazon PrimeAir once you've made it real and reliable. Until then I'm sticking with Jerry, our UPS guy. He manages to dodge trees, bushes, vegetation, poles, wires, antennae, pets, people, and vehicles almost 100% of the time and seems unlikely to defect to join the machines in the upcoming revolution against mankind. I do worry that he wears sunglasses and shorts even during the winter months, though.

Copyright Ā© 2015 Johnson-Laird Inc.