Why Drones Aren’t Ready For ‘Prime’ Time.
Two days ago I was stunned that Amazon was looking to use drones for delivery. I wasn’t amazed because it’s revolutionary, novel, or purportedly rather speedy. I was amazed because it was a huge viral marketing campaign from one of the world’s largest retailers. Also, it’s complete bullshit. Since I’m intimately familiar with utterly destroying model aircraft and subversive marketing I wanted to clarify why this is the greatest cyber Monday viral ad campaign I’ve ever seen, and nothing more. Here’s why drones aren’t ready for prime time:
Reliability
The small hex-copter drone that they show in their video looks great. Unfortunately, it’s not very reliable. As someone who’s flown all kinds of model aircraft over the years, often in high pressure situations, I can tell you that the only certainty with models is not if they’ll crash, but when. Ignore for a moment the problem of inclement weather.
These drones run on Lithium Polymer batteries. They’re a “disposable” item in the sense that they degrade just like your laptop battery. With proper care and handling you can get maybe 400 flights with a battery. If you’re willing to accept higher failure rates, you can get more. That’s about 13 deliveries per day per for a month. At that rate, each battery will need to be replaced monthly.
If the risk of a battery failing is one tenth of one percent on the batteries alone that means any given drone will fall out of the sky once every month if it’s making 13 deliveries per day. Unlike airplanes which have the ability to glide, and helicopters which can auto-rotate, when a drone loses power it becomes a very big and dangerous snowflake.
That’s just the batteries. Their are many other ‘mission critical’ components in drones that can fail. Often, when these systems fail they can start a chain reaction of failures. Take for example a speed controller. This small circuit board takes the DC voltage from the battery and transforms it into AC voltage that the three phase brushless motors use. When these fail, they tend to do so catastrophically. My critics will point out that hex and octo copter drones are capable of dealing with one or two motor failures. However if the motor’s speed control fails as a closed circuit, meaning it’s dead-shorting the battery, it’s game over.
Range
As good as modern battery technology is, it’s just not good enough for effective use in aviation. We don’t have the power density to have really effective range for delivery drones. Since quad copters are prone to single points of failure, it’s sensible to use hex or octocopters. The craft inherently weigh more, and thus require more power than a comparable quad copter. For every ounce we add in batteries and infrastructure, we subtract from cargo. Let’s assume our hex-copter is really light and we’ve managed to put a lot of batteries in it. In fact, we’ll generously assume our drone can carry 3 pounds of cargo for 8 minutes.
8 minutes of flight time at 30mph means that we have the ability to fly 4 miles. Unfortunately, the drone has to find it’s way home. That cuts our delivery radius in half. Now, you must be within 2 miles of the drone shipping center. This also assumes that there are no headwinds. In a city like San Francisco it’s not inconceivable that there could be headwinds both directions.
Oh, and at minute 9, the drone is landing on whatever it is over or plummeting like a rock on whatever is below.
A Missing Data Layer
Ever drop a pin and have an Uber show up a block away? Yeah, me too. Civilian GPS is typically accurate to about 3M. That’s pretty good, and with WiFi assisted locations, it can be better. 3M probably isn’t good enough for using drones as a delivery mechanism, but let’s assume it is.
Google maps has topographical data that’s very accurate. GPS can guide us to certain locations and we’ll know where the ground is. However, there is no data [yet] for things like power lines, trees, electric bus lines.
Imagine a husband ordering his wife’s birthday present last minute to be delivered by drone. Unknowingly, she parks the minivan in the driveway where he asked the item to be delivered. The map says it’s clear. The drone tries to land on the hood of her van and begins to make a really unsettling chopping sound as it’s destroying itself on her van.
“Damnit Josh, You forgot my birthday again didn’t you. ”
The next time you’re walking around, randomly look up. Could you navigate from 200 ft to where you’re standing very easily?
Practicality
Delivery by drone isn’t just limited by weight, it’s limited by volume. These two factors would eliminate a lot of items. A roll of paper towels might meet the weight limit but not the volume limitations. A bottle of liquid soap might fit in the bin, but it could be too heavy to deliver. These are just single items. Will Amazon send a fleet of drones to your house to deliver your Apple iPhone 8 with six additional Tsunami cables? It certainly doesn’t sound very practical.
Lastly, when these drones fail, they don’t fail in a very friendly way. Imagine a 6 pound drone coming through your windshield at 80mph. That’s going to be a bad day.
I’m an aviation enthusiast, and I think drones have an unbelievable potential to save people, protect people and improve lives, but I don’t think they’ll be delivering my Amazon Prime orders anytime soon.
Why Am I Qualified To Be Such A Hater?
I’ve built and flown radio control aircraft for the last 12 years. I have thousands of hours flying (and crashing) everything from 150CC Gas aerobats to helicopters to 13 foot Chalupa Blimps (Take that TacoCopter).
I’ve flown on prime time television, at NHL games, and during full scale air shows. I live and breathe aviation. If anyone were going to be a proponent of drones for delivery, it would be me. Unfortunately the technology just isn’t there yet.
