I can't tell, would the delivery drones be Autonomous or human piloted? This definitely sounds too sci-fi to be true. Hard to imagine the cost associated here too..
The drones would be autonomous. According to the 60 Minutes piece, it sounds like it is five years out. With a 10 mile radius, this could be an interesting proposition.
Actually I think he said the "earliest" we could expect them was 2015 because that was as soon as the FAA could approve them, thats jut a year or two away.
10 mile radius? Where are you getting that figure? Also, I thought the distribution centers were usually a bit farther out than most metro areas, making that less practical?
UAV quadcopters are kinda amazing and getting amazing-er by the year. (I spent too long looking over http://www.helipal.com this weekend.)
You can buy one for under $400 with auto-hover, GPS return-to-home, and built in cameras (or a gopro gimble attachment). Sadly, they only get about 5 to 10 minutes of flight time.
I've been checking them out too. Really cool tech. I spent a couple hours watching people's Phantom2 drones get lost/damaged on YouTube today. Convinced me when I eventually decide to get one, it won't be that model unless they've fixed its tendency to attempt a get-away.
The awesome part is, (and I'm pulling numbers out of my ass here) for a 10 dollar delivery fee, or maybe a 100 dollar subscription to prime air with an average of ~10 items shipped), this thing would pay itself off within days.
I assume that the kind of quadcopter that could lift a few kg and had decent flight time would probably run you 3-4k with today's tech, at consumer prices. Even then, a fairly good investment. At bulk prices with 4 years in the future tech, this thing would pay for itself in no time.
Good job on finding that URL. This is really cool to share with my Facebook friends, and I have to like the tone of the ad copy that Amazon is preparing on this page, which is still in testing, as it would appear.
Cool tech, but I question the practicality of it. This seems fraught with problems that a human on a bike wouldn't have. They're also entirely unsuitable for urban areas.
No, they're not entirely unsuitable for urban areas. Instead, their (relatively) short range means that they're unsuitable for anything except urban areas.
If designed poorly, they would be very unsafe, but there are plenty of ways to increase the safety. For example, notice that the drones have more than 4 rotors. This gives a bit of redundancy so that a motor can fail without causing the drone to crash.
Good point. I come from a low-rise city (Christchurch, New Zealand) so I'd forgotten about that. I suppose you could designate a safe spot, perhaps on the roof of the apartment?
I'm not sure why this keeps being mentioned. Seems easy enough to build a common landing zone for pickups. Roof, open grass area, etc. If you were an apartment owner, is there no way you could think of solving that issue?
As the regulatory and commercial frameworks evolve, I can see how buildings would have dedicated landing zones (likely on roofs)… the XXI century equivalent of mailboxes.
Can't it drop each package with a mini parachute from about roof-top level with good accuracy? Or even 10ft, just above the height of most kids, dogs, cars and adults.
Skycrane landing. I've been reading and reading and can't believe it hasn't been mentioned - the drone simply unreels the payload on a cable while hovering until it gets to the ground. Then you either detach the clips, or, use a disposable cable and just clip the wires at the drone-end (saves on possibly tangles).
The whole thing shouldn't weigh very much, assuming it's limited to delivering lightweight objects, and the rotors could be covered with a screen. I don't see much potential for danger. The worst thing that could happen would be a total system failure causing it to drop out of the sky, and even that could probably be mitigated with something like a failsafe that deploys a parachute if the speed is too high.
It should be safer than multi-ton delivery trucks, which can and have killed people. I was a mailman at one time - we were required to get out and look behind the truck before backing up because a mailman once killed a toddler who wandered behind his truck (though I doubt very many carriers actually follow that rule).
Landing on roofs? Balconies? Amazon mini-helipad mounted outside your window? :-). buildings built to accommodate drone delivery (usable by usps/fedex/ups/drugstores/pizza, food delivery?)
>>I had to seriously check to see it wasn't April 1st.
It isn't.
Honestly; I'm still not sure if this is a joke or not. This totally fits the kind of stuff Google likes to do on April 1st.
If this became real, I'd go out of my way to order things just to see that show up at my door. And I'm sure it won't be long until people plan on catching the device when it shows up. Even I probably couldn't resist the temptation of catching it and I'll probably have to pay a penalty, but it'd be worth it.
I'm pretty certain they are serious. Internally, I don't know if they really think this is going to become something right away, but I think AMZ knows that there's possibility in this field and I think they hope that this move spurs some research and progress in this area by other companies and kickstarts the needed legal structure from the government agencies in question.
If this develops more promise they will be well positioned to take advantage of it.
They'll probably include a smoke beacon/flare of some type, ostensibly to make crash sites more visible, but probably most utilized as a deterrent to would-be trappers who don't want their house to be filled with obnoxious sensory effects like smoke, odor, or alarm. Not to mention, assuming their navigation systems remain functional, it shouldn't be too hard for the drone's owner to locate and reclaim their property.
This is one of those things where I check the calendar and the URL three times each, then still don't really believe it. I think I'll check the calendar again.
It's a R&D project and the video footage shown on 60 Minutes was a combination rendering and marketing video, so we shouldn't really assume that this future is already written. You're seeing mockups.
I think it's safe to assume that Amazon is publicizing this now to help get the public on their side in any future regulatory disputes.
As a society we are going to have to figure out a lot of rules regarding drones, safety and privacy very quickly. I hope we are up to that task.
This will be a huge issue, even if Amazon gets this through the FAA, there are going to be all kinds of local interests expressing their concerns. What happens when one of these gets tied up in an overhead power lines? Did folks living near the fulfillment centers (assuming the warehouse were there first) sign up to live near a heliport, with dozens+ of aerial noise sources buzzing off every day? What about trees, etc?
This looks cool, and Amazon has some smart folks so this will likely work itself out over time, but there is going to be a lot to overcome.
It'll have to take evasive maneuvers and return fire ;) Does raise and interesting question though. Will books and other items be dropping out of the sky?
It will happen but it will also know its location and maybe video. So just like stealing on the street probably, still theft and a more risky option possibly.
I think your theory here must be the common-law view that if one owns property, one owns it from the core of the earth up to the edge of space. But that hasn't been true since the invention of the airplane: http://en.wikipedia.org/wiki/Air_rights#United_States
Brick-and-mortar stores are doomed. I can only see a few types of products remaining in brick-and-mortars: Clothes that you want to try for size, fruits and vegetables that you want to sample, and restaurants... any others?
Um, $165 isn't bad for a cashmere sweater (though to be honest I don't know anything about the quality of the stuff on this site). I took note because I don't know of a custom shirt place in Boulder, and I miss the 3 or 4 that were near my place in Houston. I hear professional women really appreciate having nice custom shirts as well.
I'm just saying, if you look at off the rack dress shirts they'll be $30-$60, similarly with most of the other items. Sure, it might be better quality, and it'll certainly fit better, but I can't afford $1600 for ten shirts. $400 sounds a lot better, and I'm not jumping up to run and get that deal either.
That's fair I suppose. I won't argue in favor of spending more just for the privilege of doing so. But A quick look at Banana Republic puts the rack price in the $60 to $125 range, and I'd hope there is a noticeable quality difference between 'fancy gap' and tailored clothing.
Correct, but hopefully they'd fit. I have a couple of yak wool sweaters from Khunu, and they run really small. Luckily my first one ended up fitting my girlfriend well enough for her to keep it. The replacement I got fits very well, and is generally amazing (as sweaters go). Now I office out of the same space as one of the co-founders too :)
If I could drone in some furniture and land it on my balcony it would open up a hell of a lot more options than having to get it up my microscopic 3 story staircase.
My parents still buy EVERYTHING in brick and mortars. I work in a retail store, selling computers and accessories. People tell me it's dead, I can guarantee you it's not. Maybe when our parents generations are gone, but not until then.
Capitalism fetishizes growth, not revenue. If the business isn't growing it will not attract investment dollars, and will quickly have a hard time staying competitive.
as long as there's free delivery/cheap delivery I would agree. However if the delivery charges are half the price of the item. No thanks, will go my ways finding the same deal in local dealers.
Edit: I forgot about pickups, I guess that counts as online. Maybe true, then
A recent trip to the Westfield Mall showed that's just about happened already. Almost every store there was selling clothing or some other product people would want to try before they buy. I you have to wonder with online delivery becoming the norm if those kinds of retail shops will become more marketing showrooms than drivers of actual sales for brands.
Wouldn't it make sense for a drone to bring you a size to try on? If it's right - you keep it ... if not, you send it back via the drone in exchange for a different size?
Brick and mortars ought to pivot into the showroom business, à la Sony or Apple stores. That is practically what many have become (cough Best buy cough); the least they can do is charge manufacturers/distributors for display space.
That implies that they can get the manufacturers to pay them. BTW, explicit manufacturer showcasing is pretty uncommon today. You have the manufacturers' stores (Apple, Microsoft) and you have some limited showcasing in large retailers like B&H for photo. But it's pretty uncommon.
Robotic sewing machines, 3D printers, downloadable textile construction packages, and motion sensors paired with modeling software to try on virtual clothes will someday end retail clothing as well.
The need to sample things by taste (or other senses) could also conceivably end in the near future through downloads of software packages consisting of instructions for chemical synthesis of tastes and appropriate output settings for the sensory simulation hardware. [1]
Just watched the 60 minutes segment: Bezos say they won't be ready until 2015 at the earliest, likely 2016. (Edit: as the commenter below pointed out, I misheard the timeframe: is is 4-5 years out)
He said the range is 10 miles, which is good enough for most residential areas. They are aiming for delivery in 30 minutes. The drones are autonomous: you plug in the GPS coordinates and away they go.
The biggest challenge (again, according to Bezos) is the redundancy, "making sure it doesn't land on people's heads."
My question is: why announce this now, 2 to 3 (or more) years out? Are they gearing up to get governmental regulations?
why not just use the delivery (UPS/Fedex/etc) trucks as a base station for the drones?
Pull in to a neighborhood of, say, 500 residents; Say you have about 100 packages to deliver... Set up 5 drones to deliver packages. Even if it takes 60 seconds on average for each drone to deliver and return, it probably is still much faster than the truck driver having to drive into every street, stop, grab the right package, drop it off, and go back to the truck and repeat.
Except that Amazon will likely offer Drones As A Service, too. They have a pretty solid track record of offering their own technologies widely, even to competitors.
I don't know where you heard 2016, he said 4-5 years, and that's being optimistic. From the transcript:
Jeff Bezos: And, you know, I don’t want anybody to think this is just around the corner. This is years of additional work from this point. But this is…
Charlie Rose: But will ‘years’ mean five, 10?
Jeff Bezos: I think, I, I am, I’m an optimist Charlie. I know it can’t be before 2015, because that’s the earliest we could get the rules from the FAA. My guess is that’s, that’s probably a little optimistic. But could it be, you know, four, five years? I think so. It will work, and it will happen, and it’s gonna be a lot of fun.
I think Bezos is wondering how society will respond. It is up to us. Some will welcome this as an improvement in delivery efficiency and service. Some will say it will make our cities crowded and noisy not only on the streets with cars and trucks, but in the sky too.
I like my blue silent sky. I recall scenes of sci-fi worlds like the 5th Element. I suggest we tax externalities. If you order by sky, you'll create visual and noise pollution to many people, and you should pay a tax for this, to be shared by the community as a whole.
My take is that Bezos wants to see how society will respond to this. A high tax would kill his ROI, and he won't invest much in R&D. If we say "Nice! We want this" he will double down.
> I like my blue silent sky. I recall scenes of sci-fi worlds like the 5th Element. I suggest we tax externalities. If you order by sky, you'll create visual and noise pollution to many people, and you should pay a tax for this, to be shared by the community as a whole.
Vernor Vinge described a similar rocket-based delivery system in his short story "Fast Times at Fairmont High", which eventually became the novel Rainbows End. Crazy to see this actually happening in the real world.
That was the first thing that came to my mind. It would be interesting to know if that was what inspired amazon or if there was some earlier source that inspired both.
Edit:
Fast Times at Fairmont High was a reworking of an earlier short story Synthetic Serendipity, which is available online:
You really want to see Jacques Tati's "Jour de fête". It's the 14th of July, French national holiday, late 1940s. There is the traditional village fair and party. The village mailman gets inspired by a combination of alcohol consumption and a stunt movie about speedy modern American mail delivery.
An absolutely hilarious prequel to this drone delivery thing, and a must see for people even remotely working on this project or anything similar.
I don't understand how this would work without having distribution centers that fully cover the nation to within 30 minutes, and having every item they sell at every such center.
Even if you just want to cut UPS out of the last 30 minute radius and not stock everything at every distribution center... that's still distribution centers everywhere.
True. I suspect that most places with a fair number of people, even in semi-rural or rural areas, in the US have a walmart within 30 minutes.
Since this service would defeat the purpose of Wal-Mart, I can't see Wal-Mart supporing it, unless there were a merger with Amazon.
But the fact that there are so many Wal-Marts suggests that it might be possible to build sufficient infrastructure that mostly covers the continental US, for example.
Nah. I'm sure they won't promise perfect availability to every customer for every product, at least not initially. There will be a gradual rollout.
But it's not as big a deal as you're making it out to be. They're already able to deliver all their things on a reliable schedule to every house in the US. That's the hard part. To do this, they just need to squeeze some more waste and latency out of the system. They've already proven to be very good at that.
Remember, our current logistics systems already move basically all the goods Americans need to within a 15 minute drive already; they just make customers do the final pickup themselves at distribution centers (which we call stores). This is just switching the last-mile burden from buyers to the seller.
If Amazon gets this working well, there will be plenty of empty big-box stores to serve as local caches.
Considering how signatures are mostly a farce (I routinely sign for my wife or my neighbours, who do the same for me), that's not a big deal: you'll just choose "classic" delivery via courier if you're really worried about that.
Interception is a real problem though, especially in high-crime areas. You could reduce it with some sort of hardened container which can be opened only with a certain code.
> Interception is a real problem though, especially in high-crime areas. You could reduce it with some sort of hardened container which can be opened only with a certain code.
Any container that is sufficiently "hardened" enough to stop any trivial access attempt would also be much heavier/larger (or have a smaller internal volume) - which would require a larger drone or a shorter distance. And short of carrying a safe, there's probably little you could do to prevent someone from getting at the goods.
Interestingly enough, the country that is likely to allow the use of delivery drones into regulations is also the country with the highest number of guns-per-capita.
> Interestingly enough, the country that is likely to allow the use of delivery drones into regulations is also the country with the highest number of guns-per-capita.
- FAA roadmap mentioning Cargo transport via drones: http://www.faa.gov/about/initiatives/uas/media/UAS_Roadmap_2....
The correlation is absurd and doesnt make any sense.
I think you've misunderstood the point of the signature.
If they drop it next door, they note which house they dropped it at and also have a signature to show someone actually took it.
The point is not to show that a particular person received the package, but that someone received the package. 99.9% of the time that's enough. If it wasn't and was impacting their bottom line, they'd have started taking photos of recipients by now.
I assume the delivery must be made available on private property to avoid the case of stealing when the drone drops the package.
As for the shooting, well it is illegal and would be punished. The copter would very probably be equipped with a video camera with a feed streaming back home which would help recognize the attacker. Then after a few publicized examples, most people would cease such activities.
It doesn't seem like the battery density exists for anything except fairly short distances with very small packages. For the vast majority of deliveries, this could only be a "last mile" mechanism, with the bulk of each delivery happening via more traditional transport. 30-minute delivery from the warehouse? Not going to be possible for 99% of purchases, even those that are under the weight limits.
Moreover, you can't just have hoardes of flying vehicles randomly over populated areas (and if this was at all practical, every company and its dog company would be trying to do the same thing), so you'd need fairly strict regulations on where they actually fly, and a system for managing them (e.g. limiting the number of flights by all companies). They'd probably need to do things like limit the paths that could be followed, maybe above public roads (which would reduce the delivery radius). Whatever happens, the (necessary, because it involves public safety and contention for a limited common resource) regulation and government control is going to slow the process down, probably a lot more than Amazon's timeline suggests...
Even if they could grind through all the problems, get government support, and setup a system to do final delivery of very small purchases by air... is the investment required to do all this worth it to be able to deliver a vanishingly small proportion of packages more quickly?
What this really looks like is an unintentional leak of Amazon's next April 1st gag...
Cars don't have eight rotor blades slapping the air, and we've gotten pretty good at enclosing all the loud bits (on expensive cars). Noise would certainly be a problem for gas-powered drones.
To minimize noise and visual pollution, it would be better for these drones to fly low, along existing roads to take advantage of existing noise abatement measures (e.g. noise barriers walls, berms, roadside vegetation, etc.)
A quadrotor (or 8-rotor here) drone depends on rapidly throttling the propeller speed on each arm up and down, something that you can't do with a mechanical transmission without adding a lot of weight. You can do a gas/electric hybrid, but again, weight.
I kind of also wonder why not a more traditional articulated rotor helicopter design. The point of the quadrotors is they're cheap and simple, despite being much less efficient. Surely the capital investment in a full fleet of these things would favor a true helicopter rotor.
Actually, contention for air space wouldn't be that bad.
Modern airplanes have a pressure transponder which announces your flight id and the air pressure at your location. From which you can determine a direction, id, and altitude.
Create an automated ATC, some simple separation rules (100' vertical, 500' horizontal), and order that drones flying east are on odd multiples of 100', westbound flights on even multiples. You know where thing is and where it's going. Maybe add in some prohibited zones around airports or tall buildings and you are pretty much done.
The thing is, if this idea proves viable, there are going to be (or at least there will be demand for) many orders of magnitude more of these than there are traditional types of air vehicles... And because of the very short range, much of that will probably be in dense urban areas (so it won't be anything like evenly distributed).
So... I dunno if traditional approaches to avoiding airspace contention will work so well...
Perhaps, but would the air traffic be any worse than road traffic?
Say you had 5 flight levels each direction, so 10 'levels' in all. Would we have more than an order of magnitude more flight traffic than road traffic? I would argue no.
Would we have more than an order of magnitude more flight traffic than road traffic? I would argue no.
We might. This sort of disruptive innovation could be staggering in scope. I could imagine putting in an order for a weekly grocery bill and having dozens of drones deliver the items from all over the place. Currently, we have an infrastructure that assumes we'll ship myriad items to a centralized clearing house (hub topology). It'd actually be far more efficient to eliminate this redundant, single point of failure and build something more akin to a mesh network. Items such as locally grown produce may even be shipped directly from farm to doorstep.
Battery density is not that much of a problem. Supply from warehouse to moving warehouse vehicle (truck?) to multiple delivery vehicles (drones). Parallellise and automate the last mile.
603 comments
[ 3.9 ms ] story [ 315 ms ] threadYou can buy one for under $400 with auto-hover, GPS return-to-home, and built in cameras (or a gopro gimble attachment). Sadly, they only get about 5 to 10 minutes of flight time.
https://www.youtube.com/results?search_query=phantom+2+flies...
I assume that the kind of quadcopter that could lift a few kg and had decent flight time would probably run you 3-4k with today's tech, at consumer prices. Even then, a fairly good investment. At bulk prices with 4 years in the future tech, this thing would pay for itself in no time.
Hope they get the FAA on board!
http://www.amazon.com/primeair
If designed poorly, they would be very unsafe, but there are plenty of ways to increase the safety. For example, notice that the drones have more than 4 rotors. This gives a bit of redundancy so that a motor can fail without causing the drone to crash.
Can't it drop each package with a mini parachute from about roof-top level with good accuracy? Or even 10ft, just above the height of most kids, dogs, cars and adults.
I'm sure there's a solution, it's incredibly fun to think about.
edit: Maybe install a landing pad by the window? People are allowed to install satelite receivers there right? Why not a small landing pad
If the drones are really good, I have a balcony they could easily land on to deliver packages right to my place in a pretty secure way.
Also, given the timeframe (30 minutes), the drone could just text me to go outside and I could pick it up on the sidewalk outside my apartment.
It should be safer than multi-ton delivery trucks, which can and have killed people. I was a mailman at one time - we were required to get out and look behind the truck before backing up because a mailman once killed a toddler who wandered behind his truck (though I doubt very many carriers actually follow that rule).
Edit: maybe not (see below).
Seriously, why the need for flight? Why not unicycle delivery drones?
The problem is, that human would insist on being paid for their labor.
Honestly; I'm still not sure if this is a joke or not. This totally fits the kind of stuff Google likes to do on April 1st.
If this became real, I'd go out of my way to order things just to see that show up at my door. And I'm sure it won't be long until people plan on catching the device when it shows up. Even I probably couldn't resist the temptation of catching it and I'll probably have to pay a penalty, but it'd be worth it.
If this develops more promise they will be well positioned to take advantage of it.
Though I can't help but think Cyber Monday eve has a lot more to do with it.
I think it's safe to assume that Amazon is publicizing this now to help get the public on their side in any future regulatory disputes.
As a society we are going to have to figure out a lot of rules regarding drones, safety and privacy very quickly. I hope we are up to that task.
This looks cool, and Amazon has some smart folks so this will likely work itself out over time, but there is going to be a lot to overcome.
And you don't own the airspace above your land, or airplanes and helicopters could not fly.
And you don't just shoot at anything in the sky when you're out shooting.
And civil courts exist to fine you heavily for this sort of behavior if it does occur.
Why don't you just shoot at UPS vehicles for driving down your neighborhood streets?
The need to sample things by taste (or other senses) could also conceivably end in the near future through downloads of software packages consisting of instructions for chemical synthesis of tastes and appropriate output settings for the sensory simulation hardware. [1]
[1] http://www.newscientist.com/article/mg22029444.500-electrode...
http://www.ted.com/talks/andreas_raptopoulos_no_roads_there_...
How do the drones deliver to apartment buildings or any address that does not have a lawn?
Would return service work the same way?
I love the product, but a little disappointed by the limited FAQ section.
I guess some of the answers related to actual Amazon policy will be clarified whenever the service actually starts (i.e. 2015 at the least).
He said the range is 10 miles, which is good enough for most residential areas. They are aiming for delivery in 30 minutes. The drones are autonomous: you plug in the GPS coordinates and away they go.
The biggest challenge (again, according to Bezos) is the redundancy, "making sure it doesn't land on people's heads."
My question is: why announce this now, 2 to 3 (or more) years out? Are they gearing up to get governmental regulations?
My guess: if the public is clearly excited, it puts more pressure on regulators to approve it.
Which is otherwise a hard sell, especially when you think of what it means for security if autonomous drones are a normal sight.
Pull in to a neighborhood of, say, 500 residents; Say you have about 100 packages to deliver... Set up 5 drones to deliver packages. Even if it takes 60 seconds on average for each drone to deliver and return, it probably is still much faster than the truck driver having to drive into every street, stop, grab the right package, drop it off, and go back to the truck and repeat.
Jeff Bezos: And, you know, I don’t want anybody to think this is just around the corner. This is years of additional work from this point. But this is…
Charlie Rose: But will ‘years’ mean five, 10?
Jeff Bezos: I think, I, I am, I’m an optimist Charlie. I know it can’t be before 2015, because that’s the earliest we could get the rules from the FAA. My guess is that’s, that’s probably a little optimistic. But could it be, you know, four, five years? I think so. It will work, and it will happen, and it’s gonna be a lot of fun.
I like my blue silent sky. I recall scenes of sci-fi worlds like the 5th Element. I suggest we tax externalities. If you order by sky, you'll create visual and noise pollution to many people, and you should pay a tax for this, to be shared by the community as a whole.
My take is that Bezos wants to see how society will respond to this. A high tax would kill his ROI, and he won't invest much in R&D. If we say "Nice! We want this" he will double down.
HA! Good luck on that.
You keep your taxes to yourself. I have a package that needs delivering in under 30 minutes by drone.
Edit:
Fast Times at Fairmont High was a reworking of an earlier short story Synthetic Serendipity, which is available online:
http://spectrum.ieee.org/computing/networks/synthetic-serend...
> "But don't worry about us." Fred looked upward, somehow prayerful and smug at the same time. "'FedEx will provide, and just in time.'"
> "Jerry was looking upward, at the FedEx shipment falling toward his outstretched hands."
Vinge also has them doing things other than delivery:
> maintenance UAVs flitted like bats around the canyons, popping out nodes here and there.
An absolutely hilarious prequel to this drone delivery thing, and a must see for people even remotely working on this project or anything similar.
But I'm old and I'll be gone someday. Y'all have fun with it.
Apparently we have very forward thinking laws in regards to UAVs
Sorry, I'm bitter still. My state is frustrating me.
Even if you just want to cut UPS out of the last 30 minute radius and not stock everything at every distribution center... that's still distribution centers everywhere.
Since this service would defeat the purpose of Wal-Mart, I can't see Wal-Mart supporing it, unless there were a merger with Amazon.
But the fact that there are so many Wal-Marts suggests that it might be possible to build sufficient infrastructure that mostly covers the continental US, for example.
But it's not as big a deal as you're making it out to be. They're already able to deliver all their things on a reliable schedule to every house in the US. That's the hard part. To do this, they just need to squeeze some more waste and latency out of the system. They've already proven to be very good at that.
Remember, our current logistics systems already move basically all the goods Americans need to within a 15 minute drive already; they just make customers do the final pickup themselves at distribution centers (which we call stores). This is just switching the last-mile burden from buyers to the seller.
If Amazon gets this working well, there will be plenty of empty big-box stores to serve as local caches.
I expect drones will increase that form of ambient noise in my neighborhood substantially.
Those who live in high vehicle traffic areas may not notice a difference.
The person who receives the package doesn't have to (or cannot) sign for it apparently.
How difficult is it to shoot down such a drone to get the loot, if you will. Might become a sport for some folks.
Interception is a real problem though, especially in high-crime areas. You could reduce it with some sort of hardened container which can be opened only with a certain code.
Any container that is sufficiently "hardened" enough to stop any trivial access attempt would also be much heavier/larger (or have a smaller internal volume) - which would require a larger drone or a shorter distance. And short of carrying a safe, there's probably little you could do to prevent someone from getting at the goods.
Interestingly enough, the country that is likely to allow the use of delivery drones into regulations is also the country with the highest number of guns-per-capita.
If they drop it next door, they note which house they dropped it at and also have a signature to show someone actually took it.
The point is not to show that a particular person received the package, but that someone received the package. 99.9% of the time that's enough. If it wasn't and was impacting their bottom line, they'd have started taking photos of recipients by now.
As for the shooting, well it is illegal and would be punished. The copter would very probably be equipped with a video camera with a feed streaming back home which would help recognize the attacker. Then after a few publicized examples, most people would cease such activities.
Moreover, you can't just have hoardes of flying vehicles randomly over populated areas (and if this was at all practical, every company and its dog company would be trying to do the same thing), so you'd need fairly strict regulations on where they actually fly, and a system for managing them (e.g. limiting the number of flights by all companies). They'd probably need to do things like limit the paths that could be followed, maybe above public roads (which would reduce the delivery radius). Whatever happens, the (necessary, because it involves public safety and contention for a limited common resource) regulation and government control is going to slow the process down, probably a lot more than Amazon's timeline suggests...
Even if they could grind through all the problems, get government support, and setup a system to do final delivery of very small purchases by air... is the investment required to do all this worth it to be able to deliver a vanishingly small proportion of packages more quickly?
What this really looks like is an unintentional leak of Amazon's next April 1st gag...
Why wouldn't these drones be gasoline powered?
… by CBS News? Don't think so.
Modern airplanes have a pressure transponder which announces your flight id and the air pressure at your location. From which you can determine a direction, id, and altitude.
Create an automated ATC, some simple separation rules (100' vertical, 500' horizontal), and order that drones flying east are on odd multiples of 100', westbound flights on even multiples. You know where thing is and where it's going. Maybe add in some prohibited zones around airports or tall buildings and you are pretty much done.
So... I dunno if traditional approaches to avoiding airspace contention will work so well...
Say you had 5 flight levels each direction, so 10 'levels' in all. Would we have more than an order of magnitude more flight traffic than road traffic? I would argue no.
We might. This sort of disruptive innovation could be staggering in scope. I could imagine putting in an order for a weekly grocery bill and having dozens of drones deliver the items from all over the place. Currently, we have an infrastructure that assumes we'll ship myriad items to a centralized clearing house (hub topology). It'd actually be far more efficient to eliminate this redundant, single point of failure and build something more akin to a mesh network. Items such as locally grown produce may even be shipped directly from farm to doorstep.