I think that the US may legislate it's self out of a really interesting industry, I would think it could be possible some smaller country with a good population density could become industry leaders here.
That sounds a lot like it will do very very well in Singapore especially if Singapore Post depots act as the hub for these drones. The entire country could be completely served by them.
Although there's no front lawn to speak of, so they will have to be able to slot things into mailboxes.
Do you have any experience with the kind of multirotor aircraft that would likely be used for short to medium distance package delivery? They are not particularly noisy, and at the distances from humans they would spend most of their time in, they probably would not be audible at all.
I have, I'm the owner of a DJIF450 and I can say they're noisy and I don't reckon they will be delivering to places like NewYork due to restrictions. If your houses are well spaced then I recon your neighbors won't mind.
I assume you're running 9 or 10 inch propellers on the F450. I suspect commercial delivery drones would use much larger props, which should be much quieter. That said, even a 9" quad is not going to be very loud 30+ feet in the air. In a populated area, especially one with cars, the noise wouldn't be audible above the ambience.
I love the idea of Amazon Prime Air, but there's one concern I have. I know next to nothing about the FAA and airspace regulation, so I'd like to hear from someone who does.
The common argument to raised concerns of drones buzzing about is "well, we have cars and delivery trucks!" That doesn't quite sit well with me because we have designated specific areas to the use of automobiles. It's easy to stay off the road. We can consciously stay out of harm's way, even if it's an asshole driving recklessly.
How is this accomplished with drones? How is the airspace structured in a way where drones are in designated areas that either avoid or are avoided by humans? How do you prevent an asshole piloting a rogue drone from dropping a heavy item on someone in a balcony? One can make the argument that these concerns also apply to planes, but I'm assuming both the airspace and ownership of planes are heavily regulated and monitored. Will we need a similar system to manage drone use?
Under 500 feet is unregulated, (FAA claims 400) the FAA is trying to regulate it since a reform law that was passed in 2012. Under 83 Feet is owned by the property owner due to some lawsuit back around WW2 involving some chickens.
I see two main options: designated virtual lanes that are programmed into drone navigational systems, and robust sense and avoid systems that allow seemingly chaotic pathing to remain reasonably safe. In practice there will probably be a combination of the two.
As for who decides what the rules are, well, in the US it's pretty easy to predict that the FAA or perhaps some yet-unconceived federal agency will try its best to strictly regulate commercial drones, and will probably largely succeed at least in crowded areas. I have my doubts that it's the best possible way to organize the activity, but it's the most likely.
I'm wondering what the hell happens when the drone gets shot down...I know if there is 100$ of good AND a multirotor then these things will get shot from the sky. No matter if they have GPS or not.
You can avoid this and all the quasi-legal unregulated airspace drama by... wait for it... just being regulated and flying higher up in regulated airspace! Sure, meeting FAA regulation is hard, but it's necessary and I'd be very confident that's the right route for a service like this. If they fly high enough the only point it can be shot down is over Amazon's depots when they pick up goods, and over the house of the recipient when it drops them off. The former is well isolated, and the latter is unpredictable from the point of view of the hypothetical shooter.
Probably, though also probably a more depressing/less glamorous version of Uber + Amazon. Having people run around in cars is going to be horrifically non-economical, and at a commodity level delivering packages is going to be far less profitable than driving people around.
More likely we'll see companies hire scores of underemployed former-members-of-the-middle class and put them on mopeds/bicycles (as density dictates). They are cheaper, and unlike machines scale up/down as seasonality requires. They are also highly flexible and capable of things like ringing doorbells, opening doors, climbing stairs, following notes on doors, and all the other details around delivery that "drones for everything!" acolytes tend to gloss over.
This is already the reality in places like New York - you can already get a huge array of things delivered within minutes by people zipping around on e-bikes.
Hehe, that's the big difference though right? We already have a really efficient, profitable delivery infrastructure. It just doesn't do near-real-time deliveries.
You can have fast deliveries, or you can have efficient/full-utilization of delivery workers. Even in a place as dense as New York City it's still very difficult to aggregate enough deliveries together that profitability for workers becomes substantial. Most are running around with 1-2 deliveries at a time.
If people want drone-based delivery, someone will make one that individuals can privately buy that will fly out to pick up packages. That's legal today. Any individual can do anything they want with a drone they own as long as it does not fly above 400 feet - it's the business owning it and using it for commercial purposes that makes it illegal under current rules.
And, like many rules the FAA makes, there is debate over whether they have the legal authority or physical power to enforce them. That's why they took Trappy to NTSB court rather than arresting him. http://www.forbes.com/sites/johngoglia/2014/04/12/faa-files-...
Where is that legal today? Don't personal drones have to be manually operated by someone on the ground within line of sight among other restrictions in the US?
The phrase “visual line of sight” is sometimes distinguished from “line of sight,” which can refer to operation in which a radio signal can be transmitted directly from an operator to a UAS that may be beyond visual line of sight.
What if Amazon are building delivery AI not only for drones, but something else (cars?)
Yeah, they say drones, but it's way too expensive project (people involved, locations, offices, etc ) to be for single purpose.
Maybe they're just building something bigger.
I'm sorry, but this is insane. No local government is going to allow these things to be buzzing around everywhere for no good reason.
People may like it for a about a week ("neato, I can get my package lickity-split!") and then will say screw this, those things are making a racket and driving me batty.
It's not like they can be used to deliver groceries. And they presumably have to return to the base after each delivery, because they can't carry that much. That's just crazy inefficient.
I could see small electric rolling vehicles that make automated deliveries. But noisy little helicopter drones? Not going to happen.
What if the base were mobile? Instead of the UPS truck driving in to your neighborhood and stopping at each house on its route, a truck parks in an out of the way corner and releases a bunch of drones to take your package just the last block or two. After it releases its fleet, the base drives on to the next neighbourhood and the drones catch up after delivering their packages to get their batteries replaced and their next package. It would increase the efficiency of the truck, the truck would still be there as a failsafe and to deliver larger packages, and most of the drone's flight time it isn't carrying a load.
The biggest challenge i see is the annoyance factor, but people have proven willing to put up with a fair bit of annoyance if it provides them with enough utility.
> I'm sorry, but this is insane. No local government is going to allow these things to be buzzing around everywhere for no good reason.
You mean like all of the cars and trucks that are rumbling around everywhere for no good reason? They roar around everywhere, are incredibly dangerous, require billions of dollars of government funded infrastructure, and destroy the environment. No one would ever allow them to become common, when we have such alternatives as walking, bikes, trains, and the like, right? They're just crazy inefficient; a bike is so much more energy efficient for a single person, and a train is faster, safer, and more efficient for moving large numbers around.
The rise of the personal car involved a huge amount of corruption to accomplish, with municipalities basically gutting the better alternatives you propose on behalf of the emerging auto-industry.
So yes, I think sans massive government corruption, we don't see these kinds of technologies that are intrusive and inefficient rising to prominence.
Also, trucks hit a technological sweet spot that I don't think drones sit in. They have predictable and easily controllable failure modes; they're always traveling at eye level where they're relatively (easier) to spot; they have relatively high transport capacity, enabling multiple visits; etc.
The problem with drones that replace trucks is that I'm not sure we've actually gained much by having flying trucks all over the place - the increase in delivery time (which I'm not sure would actually be very high in the average case) would easily be offset by the danger of many low flying trucks.
> The rise of the personal car involved a huge amount of corruption to accomplish
They may explain the US, although I doubt it. It doesn't explain literally every single other industrialized nation in the world, including the many ones without a powerful auto industry.
> They have predictable and easily controllable failure modes
So will drones, once they mature. And despite all the predictability and controllability, trucks still fall over or get themselves wedged under bridges etc, shutting down traffic for hours.
> they're always traveling at eye level where they're relatively (easier) to spot
Except when they're around a corner, and it's not much help to a cyclist about to be crushed by a turning truck.
> they have relatively high transport capacity
For drones to be competitive they will have to be competitive with this fact. If you're right, drones will never take off (pun intended). If you're wrong, you're wrong.
I think people are overestimating the danger posed by drones, and underestimating the danger posed by trucks, just because drones are exotic and trucks are commonplace. I know a family who just lost half of its members to a truck, due to no fault of their own. Trucks are far from safe.
A lot of the problems you bring up can be solved and will be, I've been working on long range multicopters so trust me I speak from experience.
At 200ft with low KV motors and big props they are nearly silent or at least quiet enough that they blend into the background. If you fly following surface streets you remove a lot of the "drone falling on someone possibility" and insurance can handle damage to cars if something were to go wrong. Keep in mind they can also be fitted with parachutes and airbags that can be deployed for safety.
Right now the biggest holdbacks are losing GPS lock (flyaways are not fun) sense and avoid (don't hit stuff) and battery life.
You can easily build a multicopter with current tech able to fly 5km+ and back in under 10min (I've done it) the issue is that due to the weight of the batteries you don't end up with much payload left to speak of. As better battery tech is rolled out and other stated problems are solved drone based logistics will be viable, I give it 5 maybe 10 years max and am very bullish on it.
Is there a reason that these use batteries rather than combustion engines?
Are you expecting the drones to use hardened GPS modules resistant to hijacking and not get DDoS'd by someone broadcasting fake GPS at their take off area?
I have been working on my quads since 2011 and have three.
Aside from their main downside (energy density), which is improving each year and now reaching 265 Wh/g, electric batteries provide many improvements. First of all, electric motors are more efficient than gas-powered motors. Secondly, gas-powered motors are fairly heavy, so you need a bigger and thus more expensive quadcopter. Thirdly, it is difficult to control a gas-powered quadcopter at the speed of an electric quadcopter.
TL;DR It's possible but difficult to make a gas-powered quadcopter, since it'll be heavy and difficult to control. At that point, it doesn't make much sense to use a quadcopter, since two of their biggest advantages are how easy they are to control and how light they can be.
Also, it would make no sense for you to "DDoS" a GPS module. DDoS = distributed denial of service. You're probably thinking of jamming (broadcasting garbage data so the GPS can't receive anything) or spoofing (broadcasting fake data.)
> Thirdly, it is difficult to control a gas-powered quadcopter at the speed of an electric quadcopter.
I disagree with you there. Collective pitch quadcopters with combustion engines have been done. With CP quads you lose one benefit of traditional quadcopters, which is their simplicity (basically, they're four electric motors with a computer that tells each one how fast to spin), but that benefit is more applicable to hobbyists than to commercial package delivery.
Ease of use, complexity and reliability. I'm referring to drones < 10kg. There's a point in large systems where gas becomes a more competitive option but for smaller multicopters it's not.
Interesting. But the real question we need to ask ourselves right now is: are these drones capable of delivering burritos? Because if so, imagine the possibilities...
Have you had any experience with or exposure to larger multirotor aircraft or any model aircraft? Noise will not be a significant concern. Carrying capacity, flight endurance, and cost are already more than adequate to make deliveries practical in an urban environment even with hobby level gear. The key missing piece right now is robust navigation software, and considering the insane pace of improvement I have witnessed just in the last two years, I am extremely optimistic.
Same argument for delivery trucks, I think what is going to determine the future of drone delivery is if it's more efficient than traditional shipping.
On the contrary, it's quite possible for practical implementation details to hold back a field for hundreds of years. My favorite example is the hydrogen fuel cell. It was invented at a time when Abraham Lincoln was busy launching his first political career and still looked like this:
yet 180 years later hydrogen fuel cells still aren't a competitive energy storage strategy. Less spectacular examples include flying cars and speech recognition / AI.
Actually, he's forecasting the future based on a misrepresentation of the present state of technology. Noise will not be a significant concern, which is obvious to anyone with any experience with larger multirotor aircraft.
