The problems Amazon faces with delivery drones

A prototype of Amazon's delivery drone service

A prototype of Amazon’s drone for delivery service

Every time I meet someone and they find out what we do at Float, they ask “So planning on getting into deliveries? I’ve heard Amazon is headed that direction!”. While it’s nice of people to think of us on the same level of capability as Amazon (maybe someday!), they’re much more brave than we are to take on this kind of task. So that’s what I want to briefly explore (in laymen’s terms), the problems Amazon is going to run into and how they might try to fix it.

The first problem really has to do with range. Basically, how far can the drone make it from the departure and reliably return. This has to do with both signal strength as well as battery life. Contrary to common thought, the battery life will likely be the more limiting factor: Ultra High Frequency (UFH) allows for a citywide distance (Team Blacksheep, a company owned by Raphael Pirker, currently holds the record at just over 27 miles), though it seems more likely that Amazon will continue to improve upon a system that uses cellular towers as antenna and would allow the drones to fly anywhere there’s reliable cell signal. The controls would simply hand off from tower to tower in the same way your cell phone doesn’t miss a beat when you’re taking on the phone while driving down the highway. Battery life typically lies in the 20-30 minute range for maximum times on most craft, but that can be significantly changed by external factors like a headwind or extreme temperatures, keeping in mind that once the product is delivered, the job is only half done. However, I wouldn’t put it past Amazon to be the player that makes significant improvements in battery life.

The second issue relates to the first: the drone’s capacity/payload. While it seems Amazon has done the research on this and understands the majority of the products being ordered on their site fit into the weight class required to go up and get an appropriate distance away with the drone (keep in mind the more weight, the less battery, the less distance). While there’s a lot of math that will need to be automated, I think the biggest issue could be the mounting of the product being delivered to the drone. Drones work best when the weight is equally balanced between the motors but just as every product will have a different weight, that weight will be distributed differently as well (think of a hammer) and could have negative effects.

The third issue for the drone services is the navigation of the craft, both flying and landing. In their most recent commercial, it appeared the end user had some sort of landing pad for the craft to identify the most appropriate landing zone, in theory making the customer responsible for finding an area that was clear of trees and powerlines etc. While the assumption is that the drones will be flying autonomously, it wouldn’t be impossible for an “air traffic controller’ of sorts to be notified at the time of a drones arrival to give a quick visual inspection and approval for landing. Navigating safely and reliably will be another test for their drones. It’s not enough for the drones to effectively deliver only 98% of the time because when something goes wrong 2% of the time there’s a lot of potential for damage to property or individuals, not to mention the drone itself. When a self driving car breaks down, it can still pull over to the side of the road. When a drone breaks down, it might not even be able to continue flying and simply fall to the ground. A drone large enough to carry what Amazon plans on carrying would be able to go through a windshield pretty easily. Liability and reliability is going to be a big player on whether these drones ever get off the ground.

So in short, Amazon has a lot of technical work ahead for it and while solving technical problems isn’t beyond a company like Amazon, solving those problems while keeping the solution light enough and small enough to fit on a delivery drone is a whole new solution. I look forward to seeing their progress, as the success of the program could be the final push making drones a common site in the American airways.