"Kimchi also laid out his thinking on how autonomous drones could safely fly in the same locations as helicopters. Helicopters are much more problematic than planes for drones because of low-altitude flying."
I wonder what size drones they are envisioning. For example with a typical prosumer drone out there now, it would be completely obliterated by a large helicopter when coming in contact with the moving blades or hitting the windshield or other parts. Likewise the downdraft of the heli would force down anything of that size in it's path.
After all if a drone is such a threat to a heli then what kind of threat is it to a person that it comes into contact with by accident?
Seems like they would need to be much larger than prosumer drones in order to carry anything bigger than a blu-ray. If a drone the size of the one in the 60 minutes demo hit helicopter blades you'd probably be looking at fatalities.
Watching the latest Mythbuster episode is enlightening, particularly if you take the weights of the crafts they're testing into account.
THe TL;DR of the episode is - carbon fiber blades directly connected to motors (likely the setup for commercial UAVs) is capable of slicing deeply enough to hit your jugular, others are not (unless the motor is being held solidly and used as a form of saw).
I'm more concerned with UAVs and payloads in excess of 10 lbs falling from 400 feet up onto my head because a wiring connection came loose and cut power to the UAV (happened to me recently, minus the head).
In short, don't fly UAVs around people, or in no-fly zones.
"A 19-year-old model helicopter enthusiast was killed Thursday when a toy helicopter he was flying struck him in the head, a law-enforcement official said."
If Amazon is going to be using these aircraft to deliver packages they are going to have to be a fair bit larger than the prosumer drones out there now to have any sort of useful payload capacity.
It will be interesting to watch how over the next decade the airspace is forced to get split up and reassigned. We're on chapter:fairy tale and we will eventually get to chapter:mundane. The tech exists, the money exists, the demand exists. There are clearly some huge obstacles, the same magnitude plaguing self driving cars. Both fascinating worlds unfurling before us.
I don't really understand this process of declaring conflicts of interests. The declaration of a conflict of interest seems like at best a bare minimum requirement and one that is abused.
I see posts sometimes where they write about how great X is and then have a disclaimer of (we're owned by creators of X) as if that was good enough. I just think if you have a conflict of interest, unless the value of the news is significant, writing about X should be avoided altogether especially if it's heavy on speculation, analysis or opinion.
That's excluding a lot of people from the conversation who might have something interesting to say. For example, are you going to exclude pilots from a conversation about aviation because they have an obvious conflict of interest?
If you're an expert in a certain area, it's quite likely that it's because it's part of your job. There aren't that many people who are both impartial and knowledgeable.
To be more specific, I am talking about journalists who work at private publishers owned by people or a company with a financial conflict of interest in a topic.
That's still preventing a lot of journalists from doing their jobs for no real reason other than a connection on an org chart. It's not like they're personally benefiting.
I think you're misunderstanding the concept of "conflict of interest". In this case, the news story was written by the same company that the news story is about. They have a direct financial stake in the public's perception of them, and are writing the story that helps shape that same perception. That's the conflict.
If the story was about American Airlines' employee compensation practices, and it was published by the Air Line Pilots Association, then the same conflict would arise. But if the story is about aviation in general, then the conflict goes away and the pilot is a good source of expert knowledge.
>For example, are you going to exclude pilots from a conversation about aviation because they have an obvious conflict of interest?
This example is not at all like the situation with Jeff Bezo, Amazon, and The Washington Post. In situations like this, there is the perception (and often the reality) that employees of the media organization would self-censor and not write anything too negative about a company owned by someone that can fire them (or indirectly hinder their careers).
> I just think if you have a conflict of interest, unless the value of the news is significant, writing about X should be avoided altogether
It's a little trickier than that: Not covering something also can be beneficial. For example, what if the Washington Post omitted coverage of bad service at Amazon?
That would ruin half the reason to own a media empire! All the big news sources are in bed with many corporations special interests, and political lobbies. The only time they have to disclose anything is if they're talking about the buying and selling of stock.
Whats surprising, is that it took something so tech-related to make you notice!
This is uncompelling. Just having layers of airspace like that doesn't really address the main issue, which is that in order to get to any layer you have to transit the layer below. Even giant commercial flights end up below 500 feet all the time of course, well at least twice for every flight.
It also has some real issues in that it's glossing over the challenges of steep terrain and obstacles. Most of our air navigation takes place by reference to MSL altitude (i.e. relative to sea level), but this is exclusively based on AGL (i.e. relative to ground level). Not an insurmountable challenge of course but it seems like they are skipping a lot of important parts. Same impression of the "real time data" plan, which sounds extremely complex in the context of aviation.
The overall impression is that nobody has really thought this through particularly, that this is more of a conversation starter at best.
The no flight under 500 ft over anything other than sparsely populated areas is in AGL, not MSL. I assume they would also need to stay out of class D/C/B areas. My main concern is that this doesn't take care of helicopter traffic that doesn't have this 500ft AGL rule.
Having ADS/B or a variant of it for the drones would take care of that issue, as in 2020 every aircraft will require ADS/B.
ADS-B also solves the tracking issue for Amazon, as each drone wouldn't need to use a separate communications channel to notify its controller of where its is; Amazon could simply tap into existing ADS-B data streams.
The positional accuracy of ADS-B as currently implemented is often measured in hundreds of meters, and GPS has never been particularly good at accurate estimations of altitude. Are drones to have ground based radar ranging? Altimeters based on barometric pressure? Updated to current altimeter settings via what means?
Anyone touting ADS-B as a realistic solution for avoiding mid-air drone collisions in congested areas is severely naive in my opinion, at least based on current technology. My assumption is that those advocating for it in this context aren't actually familiar with its specs and how it works.
In aviation, separation between flying objects is usually measured in miles horizontally, and a thousand feet or more vertically, and the systems are based on that reality for the most part.
> The positional accuracy of ADS-B as currently implemented is often measured in hundreds of meters, and GPS has never been particularly good at accurate estimations of altitude. Are drones to have ground based radar ranging? Altimeters based on barometric pressure? Updated to current altimeter settings via what means?
Some consumer drones use a radar altimeter to determine altitude AGL. The hardware is inexpensive. GPS is accurate for altitude in the US using WAAS (extremely small cost to use WAAS over traditional GPS).
> Anyone touting ADS-B as a realistic solution for avoiding mid-air drone collisions in congested areas is severely naive in my opinion, at least based on current technology. My assumption is that those advocating for it in this context aren't actually familiar with its specs and how it works.
Disagree. ADS-B was designed specifically for this circumstance (traffic avoidance).
Disclaimer: I have electrical engineering experience, a private pilot license with an IFR endorsement, and have worked on projects using ADS-B in production environments.
ADS-B just doesn't seem like a sensible solution. The only traffic you'll have is VFR, and they don't even require ADS-B out if they don't fly in controlled airspace. It seems that having ADS-B out would just clutter up the system and not be terribly useful.
I think the only viable solution would be to have sensors on the drones themselves -- visual and perhaps low-power radar. That way they should be able to avoid other drones and helicopters. It might be necessary to limit operations to VMC, although that shouldn't be much of an issue at below 500ft (you're only really going to be shut down during fog).
ADS-B being a broadcast would also notify anyone else listening to ADS-B broadcasts allowing for the potential of a golden age of drone piracy. Which would be quite profitable if you could avoid being caught.
ADS-B broadcasts contain a unique ICAO code, 6 hexidecimal characters filled in by the Transponder. The collision avoidance logic is handled by the TCAS unit on aircraft.
While new collision avoidance logic is being developed for unmanned vehicles in order to allow them to legally fly in airspace once the FAA restrictions are changed from see and avoid to sense and avoid, when it is released implementing it on an air frame will not be cheap, or light so will be reserved for larger drones. Think about 100k and 20-30 lbs in equipment for an ACAS-Xu Unit, Transponder, and antenna.
And because ADS-B is an internationally used system, changes like the addition of new messages to provide any other information commercial entities wanted their drone to provide would be unlikely to happen quickly if at all, as they would have to go through committee review.
Not nearly every aircraft will have ADS-B in 2020, as it's only required in some airspace.
Many will equip, but nowhere near all, and nowhere near enough to design a system that relies on every bogey having ADS-B out to avoid mid-air collisions.
I agree. We need to be critical of this idea because, as one of my design professors taught us, designers often "win" because they are the first ones to get their ideas out there and thus control the conversation.
I personally don't want anything dangerous or loud flying within a quarter mile of me. It's a cool hobby to do in a field, but those things in suburban or city environments could be looking into windows to see who's home, aiming bullets at kindergarten teachers or piling up C4 on a mall rooftop with next to no accountability or policing. Cars and people can do all the bad things I'm talking about, but there are license plates and cops. Looking at the frequency of attacks where non-traditional policing is the only option (e.g. online) tells us drone crime will be rampant, perpetrated by a spectrum of interests from foreign governments to financial crime syndicates, abusers and corrupt and paranoid politicians. All other dangerous networks like highways, runways and gas pipelines are tightly regulated. Drones belong in that category.
It's incredibly depressing that insane scenarios like drones piling C4 explosives on mall rooftops has become a major concern to the point where people will reject an entire category of technological devices that can otherwise massively improve (and even save) lives if used at scale.
I guess all the media and government fearmongering since 9/11 has been successful. When someone pitches a piece of tech, people now immediately think of how that tech can be used by Bad People to do Bad Things.
I would expect Hacker News of all places to not succumb to this type of irrational paranoia. But I guess we are all human in the end.
C4 on the mall is not the real issue (although it's always interesting to see how technological changes change the face of criminal behavior.)
The real issue is that I don't want a drone highway over my house. If I don't want planes flying over my house, all night and day, I don't live by an airport. If I don't want traffic noise, I don't live by a highway. But if Amazon decides that 300 feet above my house is the best route for deliveries or some other logistical exercise, I have far less recourse than I would for a road or an airport.
I think if they're only flying at 300ft, this is within the Supreme Court established zone above property in which things like drones would be considering trepassing. In some states like hopefully my own, this may also be grounds to shoot the drone down.
Piling C4 on a mall rooftop is already illegal. There is no need to make it "even more illegal" by regulating drones, and there is no hope that such a regulation can successfully deter such use, short of completely banning drones.
Cars are regulated, guns are regulated, slingshots are regulated, motorcycles are regulated, noise is regulated, EMR is regulated, medicine is regulated, knives are regulated, RC airplanes are regulated. Anything that can cause danger is regulated. Leaving drones unregulated is absurd.
There is tremendous likelihood that regulating drones will lead to less crime. Banning drones within 1/4 mile of permanent residences and mall rooftops will prevent a large chunk of drone crime.
Banning drones from mall areas is not about "making it more illegal to put C4 there." It's about empowering the law and private citizens to forcibly remove drones from a dangerous areas to prevent drone crime.
If drones are allowed to buzz around malls, there are no police to determine if one of them is piling up C4 on the roof. If drones are not allowed to buzz around malls, as soon as one as spotted in a no-fly zone, it can be taken out to prevent a dangerous crime.
If you enforce the regulation with automated anti-drone guns, then yes it will make a difference. I don't think it'll happen though, as such countermeasures haven't been developed for existing security threats.
There's an awful lot of stuff below 500' AGL - buildings, power lines, towers, etc. Those are required to be marked and illuminated near airports, but not elsewhere. Near airports, controlled airspace extends right down to the ground surface, and typically five miles out from the airport. That covers big chunks of many cities.
I just looked at the live map for the SF area. A Temporary Flight Restriction over most of SF begins in 39 minutes, surface to 3000 feet, because there is a baseball game at the ballpark.
Autonomous drone operations in urban areas are going to be very tough to control. Drones are going to need their own airborne radar. Maybe only a hundred meters of range, but they're going to need to be able to see details down to power lines. This is available now, but needs more downsizing.[1]
What might work is a system where all manned aircraft have ADS-B (this is coming) and drones also have ADS-B and enough smarts to stay well out of the way of anything bigger. Drones also need short-range radar so they can avoid bumping into obstacles and each other. Think of this as "micro-see-and-be-seen'. It's a lot of electronics to carry, but less electronics than a smartphone.
This is more likely to work than some scheme where everybody phones home to Amazon's cloud.
First I recognize you said maybe, but the range would need to be much larger than 100 meters.
As out of the box hobbyist drones can already travel at 50 mph [1] (or more I assume). And 100 meters assuming approach of a stationary object would provide just over 4 seconds for the drone to respond and avoid the object.
While 4 seconds is sufficient time, it will lead to drones flying very close to objects before avoiding them which will not inspire public confidence. Also as faster drones appear that window will become even smaller. They should probably have radar with a range measured in nautical miles to avoid having to implement a new system almost immediately. I would say at least 5 nautical miles, although I feel even that would be on the short sighted end of implementation.
They've been clocked at 80+. Remember though, they're not going to be descending at 80mph, which is really the only time you'd need to worry about collisions.
That is definitely not the only time you need to worry about collisions, it is the most dangerous time but not the only time, as there are many things taller than 200 ft, as well as birds and unequipped vehicles using the airspace in an unauthorized manner for you to worry about.
Radar power needed increases as the fourth power of the range.
The state of the art in radar is that sensing power lines and similar wires is not reliable. 77GHz radars employed in cars can sense people and bicycles, but that's not good enough. There's been much effort put into power line detection for aircraft (helicopters hit power lines about 50 times a year in the US) but no good solution yet. Drones zipping around streets with overhead wires and trees is not yet workable.
It seems like power lines would be able to be detected with system measuring changes in magnetic fields, perhaps based off current induction sensors. However clearly a system like that does not yet exist and that would not help with other obstacles. And I agree with your statement that this is currently not workable.
My personal idea for this to happen is that every drone should behave as a giant beach ball, physically. The drone would be in the hypothetical center of the ball, and it can move around freely as long as "its ball" does not hit other balls. When it does encounter other balls, the drone performs a physics simulation of a beach ball, making sure that the center of the beach ball will never hit the center of another ball by taking into account "compression" forces, etc.
It never ceases to amaze me how negative the HN community is to new ideas, technology and generally "trying new stuff" that has never been attempted before.
Drone Delivery - never going to work. Lots of solar power - too expensive. Autonomous cars - too risky
This is supposed to be a community of "Hackers" making cool stuff, and here we have a company talking about making cool stuff, and every single comment, without exception, is negative.
Yes, this is a very difficult problem to solve. Yes, there are many things they have not addressed yet. Yes, this needs a lot of work. No, they shouldn't just start doing this and see how many crashes happen. But yes, they need to keep talking, testing, learning and working towards a future where this is possible.
Does anyone in their right mind honestly think that in less than 150 years there will still be a single human-controlled vehicle (plane or automobile)?
Of course it's coming, and these guys are starting it. Now.
The same negativity could be said of the Wright brothers when they told people they were going to fly, or Henry Ford when he set out to mass-produce the automobile, or the people that said we'd have a global computer network and a computer in every home.
I honestly don't understand why the community here isn't more open to new ideas. Even ideas that never come to fruition are still worth analyzing. There is always something to be learned.
We should be talking about the interesting parts of this challenge and how they could be overcome, not saying "Not safe enough" Every. Single. Time.
"Does anyone in their right mind honestly think that in less than 150 years there will still be a single human-controlled vehicle (plane or automobile)?"
You do realize that we don't even have clean water on a global scale?
Right, but still only twleve of us have been to the moon.
I think the parent commenter's point was that it's pretty naive to think that there will be no more human-driven vehicles at all. Change will happen quickly in some parts of the world, and other parts will take a long, long time to catch up.
Yeah, so we spent a trillion dollars sending a dozen guys to the moon because people in Kansas were shitting themselves thinking about Soviet space superiority. Now tell me about the dozens of assholes who said we'd be terraforming Mars by now, or be flying between Tokyo and LA in 2 hours on hypersonic planes. Heck, tell me about DRAM latency since 1995, the fact that power generation is still dominated by coal plants that are less than 2x more efficient than they were a century ago, or the fact that the sexiest thing in aerospace right now is a company trying to get back to where we were in the 1960's.
These are two pretty unrelated issues, and it is a fallacy to compare the two in this context. It's not unlikely that the majority of transit in the western will be automated in the next century or two.
The technology to filter and clean water has existed since the Roman era, but its implementation is highly dependent on political and economic factors (fun fact: western civilization didn't match the Romans' quality of water engineering until the mid-19th century).
The technology for automated cars could well exist in 150 years, and its implementation depends on the cycles of industrial production and consumption. It's likely that much of the western world will have automated transit in 150 years, because everyone will eventually get rid of their 2015 toyota for a model 2077 automated Tesla-google car.
Sadly, there's not as much financial incentive to get clean water to people often without the means to buy it. In places often wrapped up in corruption, war, etc.
With automated transport, the early movers could absolutely dominate the future of transport.
And it never ceases to amaze me how the tech community so easily blows off legitimate safety concerns.
You can't just "reboot everything" after a drone wipes out a Cessna or smashes a person's skull. Any drone capable of delivery is going to be big, powerful and not light.
>And it never ceases to amaze me how the tech community so easily blows off legitimate safety concerns.
I seriously doubt any drone proponent will be dismissive of safety concerns. If there are people who don't care about the safety concerns drones bring, this is just the other side of the same coin with people so resistant to the ideas of autonomous vehicles. Both are people not forward looking in their own way.
When Amazon says, "hey we want to use drones," reactions shouldn't be, "hell no there's no way you can make that safe," and instead, "great idea, let's make sure there is a discussion at large to reach a consensus on the safety standards required to make that happen."
Big companies very consistently skimp on things that are not directly related to the success of some undertaking. Sure, safety standards might be put into place, but that does not convince me in the slightest that Amazon is going to spend a lot of money making sure the drones are secure and bug free and can't be hacked easily, so on and so forth... Certainly not to the degree that commercial airplanes are vetted, and even those systems aren't invulnerable.
> This is supposed to be a community of "Hackers" making cool stuff, and here we have a company talking about making cool stuff, and every single comment, without exception, is negative.
Most engineers are conservative and skeptical. If anything, as someone who is an aerospace engineer by education, I'm always surprised by the wild eyed optimism around here.
> The same negativity could be said of the Wright brothers when they told people they were going to fly
It could also have been said of the generations of idiots that jumped off cliffs to their deaths in flying machines that had no hope of working.
In any case, I don't think people are pessimistic about the technology. They're pessimistic about the ability of our social institutions to accommodate and manage the effects of that technology.
Identifying a problem is the first step to solving it. And there's far more likely to be people willing and able to take that first step than any subsequent one.
>It never ceases to amaze me how negative the HN community is to new ideas, technology and generally "trying new stuff" that has never been attempted before.
That's because the HN community has a high bar, and most "new ideas" are bullshit (when it's not a rehash of failed old ideas shot down 100,000s of times, like "perpetual motion machines").
> It never ceases to amaze me how negative the HN community is to new ideas, technology and generally "trying new stuff" that has never been attempted before.
Partly I think this is people setting themselves up to be able to say "I told you so!" at a later time. You look less silly predicting a failure that doesn't happen than if you extol the virtues of an idea that later turns out to be complete hogswash under further scrutiny.
Though to put a more positive spin on it: critical thinking is a much neglected skill. Sometimes things genuinely aren't well thought out, and identifying and documenting a problem are the first steps to someone coming up with a solution.
I'm sure there are more elegant ways to address this, but couldn't regulators just stick a drone no-fly zone where piloted aircraft could be expected to be below 500 feet?
Since there will be so many flying at once you're going to have the occasional failure. That's going to make walking around in the city interesting for sure.
It should be straightforward to design corridors for drones where, if they were to fail, they land on structures instead of people. Amazon could design this as an additional OpenStreetMap layer, using existing satellite mapping data to create areas to avoid where people are likely to be.
You know even if they don't land on people, I sure don't want an Amazon drone falling on my house, or out in my pasture, or on top of my shop. Will Amazon pay up in full if it punches a hole in my barn, harms livestock, or destroys a car I have sitting outside? If they're going to have a non-zero number of these drop from the sky, they better know they're not getting that drone back, for starters.
First, its no different then an aircraft. You know what happens if a piece of an aircraft lands on your house? The airline's insurance company pays for it. You know how much say you have over airspace above your home? Zero.
> If they're going to have a non-zero number of these drop from the sky, they better know they're not getting that drone back, for starters.
While you may think property rights don't exist solely because of the word "drone", you'd be mistaken.
I have serious doubts you're going to see significant drone traffic in the area where you live (based on your profile info).
The NY Post is going to have a real selection of headlines when some drone crashes into some poor person walking down the street. Depending on the cargo, I expect they will rise to their tradition[1].
No worries, we will just build other drones that hover at 100 feet above ground level, and are programmed to detect falling drones or drone parts, and then cast a net to catch the falling parts. Then, we will have other drones, that hover at 40 feet above ground level and are programmed to detect falling drones responsible for catching the parts at 100 feet, and...
Agreed, there needs to be rigorous regulation on noise pollution, especially at the expected high densities these devices. In contrast, autonomous electric vehicles are quiet and can handle much higher loads and many more loads.
Has there been much discussion about hundreds or thousands of drones filling the sky and how much of an eyesore that would likely be to many people, then the political fight that is likely to ensue?
I still don't understand how an autonomous delivery drone is better than an autonomous delivery car.
I feel like a delivery drone is the absolute most complicated way to solve this problem other than somehow involving rockets.
Drone limitations:
- Needs great weather.
- Package size and weight exceptionally confined.
- Exceptionally limited range.
- How to actually get package from drone to human? Drop?
- Night flight? Yes / no? Probably not.
Using delivery cars would cause a lot of traffic in densely populated areas. Cars are presumably also more expensive than drones. On the other hand, I suppose a car could do more than one delivery at a time whereas a drone would be more limited in what it could carry.
This summarizes a big problem with the congestion argument. Our economy is not big enough to cause "serious" delivery congestion when confined to 2-d road vehicles, yet its assumed if you open up to smaller 3-d capable vehicles massive congestion will result. Why?
Where I live it is not an issue, but it is theoretically possible that there might be a location on the planet right now where Fedex trucks crash into UPS trucks, and Dominos pizza cars are running Chinese delivery cars off the road, on a daily basis. However that weird situation existing would still be a workable system for perhaps 99.9% of the worlds surface.
There are also practical scaling issues. The local pizza delivery joint cannot produce fast enough to saturate a 2-D delivery system during the superbowl, just can't pump out pies fast enough. Naturally going to 3-D delivery with maybe 100x the delivery capacity because of 3-D isn't going to alter their delivery rate, the skies will be pretty empty. Even if you drive there to pick up, there will be a 2-hr wait for pizza. Drones and trucks are expensive and the finance guys would prefer the absolute minimum be owned and deliveries be made 24x7. That would imply we're not going to get drone congestion until we have deliveries at 3am. I'm just not seeing people order enough raw "stuff" to keep delivery services busy enough.
How much more traffic would there be if each delivery service had to drive back to their distribution point because they could only deliver one package at a time?
Actually drone delivery is much more simple than autonomous driving. Algorithm: grab package, fly up, fly directly to delivery point, drop package, all faster than driving.
Compare to autonomous driving which needs to take into account road users (non-autonomous), traffic laws, visualization, etc and it's obvious why drones are more efficient and easier to program.
Assuming empty airspace, it's obvious why drones are easier to program. Assuming a single package it's obvious why drones are more efficient. Take away either of those assumptions and things get less obvious.
Those are very wrong assumptions. If a serious drone delivery program (or multiple competing ones) starts, airspace will suddenly become congested and you'll have to do heavy air traffic coordination to avoid mid-air collisions. And when you comapre with cars, you need to compare one delivery car with a hundred drones. Suddenly you see that drones are not efficient at all, which is kind of obvious, as it takes a lot of energy to keep stuff in the air with active propulsion.
I always pictured trucks driving from distribution centers, and deploying drones that cover an area. That way the drones only have to cover a small distance, reducing the number of drones that have to fly over a given area.
Mobile drone hubs seems like a very interesting solution, which could even be more energy-efficient than just trucks or just drones. Somehow I never thought of that, thanks!
Airspace is big. A central registry where drones announce their flight plans, allowing other drones to selfishly avoid them, would allow for plenty of room for drones to make sure they stay out of each other's way.
I'm skeptical on drones, and I think all this is an effort by Amazon to get in the news. But this doesn't seem like the problem.
Sort of, but as long as all drones (in this airspace) are automated, it's still an easier problem than self-driving cars. Just detecting where a road is is a feat, and then throw in other traffic, and then say that traffic is driven by people, who tend to behave irrationally by a computer's/programmer's typical definitions of "rational". Then there are traffic controls, construction, and the like.
Amazon's idea is to have airspace dedicated to autonomous drones, who can go from a to b in a straight line and mostly just have to worry about not hitting other drones (or buildings or birds or what have you). You can do that with some decent sensors and some simple physics simulations.
That's not to say that airspace is totally unregulated, but the regulations tend to be much broader (AFAIK; I'm not a pilot).
Collision avoidance is far easier in three dimensions than in two. Those extra degrees of freedom are crucial. (Of course, traffic capacity is much higher.)
It is only the severity of an accident that increases, not the likeliness.
I imagine it's because airspace (at least up to 400/500ft) can be relatively empty compared to road traffic in large cities which can be very congested (especially at peak times). So it could lead to reduced delivery times as well as potentially being cheaper than cars. This depends on exactly what the drones are capable of delivering (not your grand piano, of course).
No traffic until people actually start using drone delivery systems. It's like with planes, the air seem so big but in fact it's very crowded and you get mid-air collisions all the time (hell, one case was just being discussed on HN earlier today). Scale that up to account for more drones in smaller space.
Apart from that, it still isn't obvious to me how this is better than an electric delivery truck. Cars are much more energy efficient than drones (especially helicopter-like ones), doubly so if you load the car to its capacity. Note that in case of drones, the mass of the package is only a small percentage of the overall mass of the vehicle, so most of the energy used will scale up with number of drone deliveries. I'm pretty sure taken altogether cars will win in terms of energy use.
Depends on the type of drones you want to use and their range. Helicopter-like drones have very poor range and flight time, so this would limit your range considerably. And if you go to a plane-like drone, or even some kind of hybrid mode (eg. VTOL), suddenly you can't really do an emergency hover when cruising to destination.
But even if they find a way to handle the traffic, I was just saying that there will be traffic and that this poses serious problems.
I think the traffic Amazon would be worried about is human drivers, not other drones. AFAIK, Google's driverless cars have been in 14 documented accidents and all 14 were the fault of human drivers (including one incident of a human driving the Google car). Machines may be better drivers than humans.
Will drones or self-driving trucks include Magic Doorbell-Tropic Technology[tm]?
How will they put deliveries into letterboxes? Or will there be a special Drone Nest Drop Off Utility Safebox[tm]?
And so on. The more you think about the idea, the less practical it gets.
I'm sure the problems can be solved. Obviously, you could retrofit your doorbell to include a drone code. Or make an app (it's always an app...) that alerts you when the drone is near.
This is perfectly fine, unless you don't want to change your doorbell and/or don't have the right kind of phone and/or don't want your stuff stolen.
There's a whiff of "Reinventing Yesterday's Technology for Tomorrow" about the idea. It looks a lot like a high school science show-and-tell trying to turn into a delivery system that can compete with 75 years of professional logistics expertise.
I don't usually do "not invented here", but there are so many very obvious things that can go wrong, and so may ways users need to change what they do, that I think debates about air traffic control are a long way down the list of show-stopper issues that need (literally) bullet-proof solutions.
I would like to see a startup for what amounts to background checked, bonded, doorman service. Someone in my neighborhood I could trust to take deliveries, stuff like that. Outsourcing trust, I guess.
I could see drone delivery company "tipping" some old retired guy to solve their "last tenth of a mile" problem as he walks down the block to my house.
Alternately make all uber (or insert other) driver watch some training videos and tell them to be at xyz address at abc time and they verify safety and walk the last 50 feet and hand deliver, for a modest tip of course.
Presumably if I watched the training videos and didn't have badminton nets and kids pool and ham radio antennae in my backyard the drone could deliver directly to me and save me the cost.
For one, you could always have a human operator coordinate the landing and delivery process. They could coordinate delivery using a phone call.
Secondly, you could have a drone deliver goods to a distribution truck (possibly with a special landing-pad marker on top,) and a human could then deliver the package manually.
No advanced technology needed for either of these, it's just communication + drones.
I believe the ultimate solution will be a combination of the two:
You have an automated road vehicle that acts as an "aircraft carrier" for drones. Something roughly the same size as a delivery truck today. It gets loaded up at the distribution center with packages and a few drones.
It picks a route that gets roughly near all of the deliveray locations. As it goes by, a drone is dispatched to take the package a couple of blocks to the person's doorstep and then returns to the "aircraft carrier".
When the drone gets back to the truck, it plugs in an recharges.
This uses drones for what they're best for: quickly and efficiently navigating small complex spaces like side streets and front yards when there is only a small payload you need to deliver. The truck addresses the shortcomings: limited flight time and payload capacity.
At this stage, I think it may be easier to automate a drone than a car. Buildings and hills don't move around that much, and air traffic won't be as random as roadway hazards if implemented correctly.
You could have said the same thing about horse drawn carriages in comparison to walking. Need to keep the horse healthy, need to keep the carriage working. Walking: need to be alive. At some point, we make sacrifices for convenience's sake.
“Imagine the Internet without HTTP and TCP/IP” That's quite an absurd analogy to push forward Amazon's drone program.
Even though I am quite fascinated by the innovation of this idea, not so much about the practicality of it. I believe it is good for rural areas where highway system is not streamlined and major shipping carriers don't encourage frequent routes.
I like the idea of rivers serving as a natural drone highway. They weave through most cities/towns, provide insulation from populations on the ground and relatively easy recoverability (as opposed to being destroyed on impact).
Also consider railroad tracks, at least normally you'd only have one property owner making claims (the railroad), and the local residents are already used to 9000 HP worth of coal train rumbling thru at 3am so a tiny little quiet drone is no real issue. Also in an accident, I think its pretty obvious who "wins" when a ten million pound train hits a ten pound drone, so other than glass and paint claims I think you're pretty safe. I don't envision a drone ever causing a coal train to derail, LOL.
>While the Supreme Court hasn't explicitly accepted that as the upper limit of property ownership, it's a useful guideline in trespass cases. Therefore, unless you own some very tall buildings, your private airspace probably ends somewhere between 80 and 500 feet above the ground.
It would be interesting how they plan on detecting smaller moving objects like birds, which would be more of a challenge than a helicopter or a construction crane. Perhaps lasers and radar similar to how Google's driverless cars detect potential hazards?
I wonder if detecting them is even really a problem. They move, emit heat, and most are naturally inclined to avoid drones already. I'm really curious about whether birds of prey might start hunting drones.
More fun will be in optimizing stopping birds going for drones. My little photo quad has been attacked by mobs of both seagulls and crows. Wasp-coloured paint and ultrasound squeakers?
They didn't address the other part of safety -- some percentage of those hundreds (or thousands) of 15+ kg drones flying over a city are going to have an unavoidable collision or a catastrophic hardware failure and come crashing down to the ground, which could easily be deadly if it lands on someone, or cause a car accident if it lands on a car or on the road. At the very least it's going to cause property damage - who do I call when one punches a hole through my roof or I see one hanging in my tree?
Has anyone done an analysis of the crash rate for large drones and likely casualty rate on the ground?
Well, you could always take anything onboard the drone as salvage. Or what about all those people who will not want to let the drone retrieval squads onto their land?
I'm waiting for the ability to transmit "bird alerts" to chase away the drones I don't like.
You're right that they need to address those concerns. I'd like to see that analysis also include any possible reduced injury/death from fewer delivery truck miles and fewer consumer car trips to the local store.
I have no idea if it will offset or not, but I could see—even with a few thousand drones crashing every year—that a net benefit could still be realized because people are not driving to the store for some of their shopping.
Sure, using drones instead of trucks could be a net win in safety if it means less drivers on the road, but this seems unlikely -- one driver in a UPS truck can deliver hundreds of packages on his route, but since the drone has such limited payload capacity, it might need to make a round trip per package so it would take hundreds of drone trips to replace one driver with a truck. And, since heavy packages aren't realistic for a drone, the truck is still going to drive much of his delivery route to deliver the heavy packages.
I think setting up a regular UPS truck with a couple of drones that land on the roof to pick up small packages and deliver them is pretty compelling.
In Houston there are only a couple of logistic centers for UPS and FedEx and whatnot. You wouldn't have a drone fly 40 (or maybe 100) miles round-trip to deliver a package all the way across town.
But if every regular delivery truck has 2x or 3x drones on top they can make small package deliveries without the driver stopping on his route as he/she delivers the bigger packages. The ones too heavy for a drone to lift.
The drones could make deliveries during the driver's lunch break which is a nice bonus. An hour with the truck parked in the same spot.
You could also have drone-specific semis that go on a very slow rotation around the city, moving a few miles quickly and then stopping so the drones can make deliveries from a "home base".
Drones are a good last mile solution but will never replace the back haul network. Wheel on pavement has incredible efficiency versus rotor through air.
> the truck is still going to drive much of his delivery route to deliver the heavy packages
How this will exist in the future in practice is pure speculation at this time. There are many different factors (autonomous vehicles, smarter routing techniques, routing changes as the burden is shifted to non-truck delivery systems, and increased demand for small package delivery as more shopping is done online, for example) that could dramatically reshape the delivery landscape.
Furthermore, heavy packages delivered to consumers in the US, like those containing appliances and furniture, are already routinely delivered by in-house or smaller delivery services rather than UPS, FedEx, or US Mail.
What if a drone costs 100x less than a delivery driver? You can have 100 drones out delivery packages for the same cost of one delivery driver delivering one package at a time.
It would be interesting to see data on what % of amazon purchases are above the payload of a drone - i imagine they crunched the numbers themselves and determined it was low enough to justify continuing to push this issue (even after the FAA tried to ban[1] drone deliveries altogether)
Theyll look at $/package and time/package for drones and delivery drivers.
I imagine once you account for driver salary and benefits, cost of vehicles, cost of fuel and maintenance, cost of insurance for drivers and vehicles, cost of management structure to maintain drivers, schedules, routes, etc - the $/package for drones likely make up for any disadvantage they currently have in time/package
Also, drones can deliver packages 24/7/365, no reason you cant wake up to a package delivered overnight (maybe even your morning kindle (tm) e-ink newspaper)
As drones and other autonomous machine/robots start to replace human employees, I expect to hear calls to tax them to recoup the social externalities. If they are taxed enough, then there is no longer an advantage to using them over human employees.
The fallacy in that thinking is left as an exercise for the reader.
24/7/365 seems overly optimistic, weather alone will most likely cut out many of those trips. Depending on the type of collision avoidance technology used may further limit night time operation.
UPS-type drivers are very safe. There are lots of trucks, resulting in many accidents nationwide, but they are far safer than the average driver. (Remember that many/most road deaths occur at night and have some relationship to alcohol.) So taking some of them off the road won't contribute much compared to the slight but undeniable risk of a drone-aircraft collision killing many.
Yeah, they have tremendous safety records. I should have focused on the other "opportunity cost" of fewer deaths dues to fewer car trips made by consumers.
Also, we have no idea what the drone-aircraft collision risk is. That seems like something that could be easily solved by strict control over who flies where in the sky. I'd guess that more people will die from these things falling on their head.
Drones can be engineered to be far more resilient to failure than what we are currently seeing today.
The DJI Phantom for example only has 4 rotors and a single battery, so any one failure of a component in the power train or controller will likely bring down the entire device.
The type of drone Amazon is likely to deploy widely in this scenario would of course be much larger, but would also have many more rotors, think of a copter with 8 arms and 2 motors per arm, with the power trains wired up as 4 independent systems each with 4 opposing rotors and it's own battery capable of keeping the device flying independently. Flight controllers can also be built hilariously cheap, so having 3 or 5 in a voting configuration is equally simple. Now you have a machine that can stay airborne and under control with over half of the rotors disabled.
Sure, engineering can reduce the risk, but even passenger aircraft sometimes suffer catastrophic failures and drones will likely be under much less scrutiny than passenger aircraft - $10,000 metal fatigue scanning may make sense on a $3M aircraft, but on a $10,000 drone? what will happen to your 8 armed aircraft if the batteries burst into flames, metal fatigue in the central hub makes 3 of the arms on one side fall off, or a rogue drone runs into it?
When drones are making millions of flights a year, even a one in a million malfunction is likely to occur many times a year.
Cars kill ~38,000 people per year just in the US. Considering the number of cars on the road would go down (used for deliveries or trips to the store) The number of lives saved would most likely counteract any lost to the 1 in a million chance of a drone falling and killing you. Zero risk is not going to happen but given enough care is provided and sane regulations are in place drones should have a net positive impact on the number of people killed each year.
It's a shame we have become so risk adverse as there are plenty of things way more likely to kill you then drones.
All I'm asking is to show me the numbers and quantify that risk. With dozens of drones flying around there's minimal risk, but what's the risk look like when there are many hundreds or thousands of drone flights every day across every city? People and buildings have already been hit by drones despite the fact that drones are relatively rare.
There as well over 1 million drones currently in the US. I get that number by adding up sales figures or estimates of the bigger manufacturers. (Parrot, DJI, 3DR etc..) There could certainly be well more then that (and most likely are) as that's not including hobbyist DIY drones. I've yet to find more then a handful of references to deaths by drones over the past few years and even then those were caused by Single Rotor RC helicopters or RC planes (if you want to lump those in as "drones")
While I agree with you, I think people are concerned because they know how to deal with a car accident. However, as one user asked before, who do you call when the drone crashes into your house? What about your child? No one is really proposing an established procedure and quantifying the risk yet, which is scary.
Regulators could address this concern by requiring drones to carry identifying documents (e.g., license plates) and liability insurance. This model has worked for truck deliveries for a century.
It's not scary to me. Who do you call when a drone crashes into you house? You call the police, obviously, or emergency services (911) if someone is hurt. Who else would you conceivably call?
Police: "Did it cause more than £2000 worth of damage?"
Me: "I don't think so."
Police: "Call your insurer. This is not a police matter. Bye."
Insurers will presumably say "that's not covered" up until the point where they have an extra charge for drone insurance when they can say "you don't pay for that cover".
I have no idea if that's a realistic hypothetical. Why would drones be any different than something else causing minor damage to your property? What do police do if a kid hits a baseball through your window? I'm sorry, I just can't work up the existential fear of drones that people express.
Where I live, if a drone crashed in flames, it would start a fire that would replay the Oakland Hills fire [1] with people getting killed. If a car were to crash on Sand Hill Road, not so much.
That's pure FUD, a car could easily cause the same fire, and the average fire in the hills does turn into a firestorm, it requires a particular set of circumstances.
- Cars travel defined roadways having mandated brush clearance zones that diminish the likelihood of a vehicle crash/fire propagating into uncleared country. A fiery crash is very likely to be noticed and reported immediately, with no uncertainty about its location. Fire responders will be quickly dispatched to the correct location. The road on which the crash occurred will very likely give responders good access for their suppression work. If evacuation is needed, it's easier to direct evacuees to routes that will be most effective in getting them out of danger.
- A drone isn't constrained to defined roads. Compared to the car crash scenario, when a drone crashes away from roads and ignites vegetation, more time will pass before the fire is noticed - reported - located - and fire suppression resources put on site. Away from roads, this means airborne fire suppression and that adds at least 30 minutes to the response loop, even when (as now) there are airborne assets on alert. So an off-road ignition is almost guaranteed to have a longer incident response time than a car crash. Same is true of other off-road fire events like careless hiker cigarette toss, camping fires, etc. Directing evacuees which way to go becomes more difficult.
- Favorable firestorm circumstances: even small fires can combine with topography to generate fast up-slope winds that, while local, are very effective in broadcasting burning material that ignites locations remote from the original burn.
Summarizing, if a drone crash starts an initially unobserved fire in an uncertain location that's tough to access, there's a larger chance that it will become a danger to the local rural population than one started by a car crash.
It's FUD because you're citing a worst-case scenario. I'll acknowledge that a drone is more likely to crash in a dangerous fashion (from a fire perspective), but it's still insignificant compared to the real risk that we live with every day from automobiles.
> drones will likely be under much less scrutiny than passenger aircraft - $10,000 metal fatigue scanning may make sense on a $3M aircraft, but on a $10,000 drone?
I'd like to think that the cost of a passenger aircraft isn't the only reason why they come under more safety and regulatory scrutiny than a motor car. I'd hope that it's because they have an additional severe failure mode that cars don't called falling out of the sky.
I realise that I might be going on about this and I'm a layman, but it seems a bit overlooked that any object that is airborne and needs to be made safe surely starts at a large handicap compared to objects on the ground.
So just like my car insurance, I'll have to purchase "uninsured drone operator" insurance to cover my house when a uninsured or unidentifiable drone drops on my house?
An uninsured aircraft falling on the roof would surely be a valid claim. And anyway companies like Amazon would surely be required to have their own insurance.
Read the link I posted -- if you make multiple claims (even if they are all perfectly valid claims and not your fault), your homeowners insurance rates can raise, or you can even be blocked from coverage from multiple insurance companies through the industry wide C.L.U.E. claim reporting system.
Sure, if it's a known drone from a large company, I'm sure they will have insurance, but what do you do when you have a $5000 hole in your roof and Amazon claims that the "Amazon" drone that you found is not theirs - someone slapped an Amazon label on their own drone to get away from the insurance requirements.
I'm sure your existing homeowner's insurance policy will cover that. I'm certain of it. It may add an extra $0.50 a year to your premium, because the odds are so small that:
(a) Your house will get hit by a drone
(b) The collision will cause damage to the building
(c) The drone will have no ownership identifiers on it.
(Hint: Not Amazon, which is what we're discussing here)
You don't currently purchase "uninsured baseball thrower" insurance, do you? Those things cost tens of millions of dollars of damage in broken windows every year, but we've figured it out.
My house has already been hit by a drone, albeit a small $500 drone, though there was no damage. And the owner asked if he could retrieve it so I lent him my ladder.
But as drones become more ubiquitous and the big ones become cheaper and more useful, then the chance of damage increases.
Amazon isn't just talking about legitimizing their drones, but legitimizing drones in general with an "anything goes" airspace below 200 ft.
I wouldn't make a homeowners insurance claim for a baseball through the window since claims mean higher rates and multiple claims can lead to blacklisting among multiple companies. But a 20 or 30 lb drone dropping 500 ft to my roof may cause thousands of dollars of damage, and unless I can identify the drone's owner (or even if I do and he has no money), then either I or my insurance company is on the hook for repairs.
Regarding the baseball. Usually, if the miscreant that threw it then ran away can't be identified, you just eat the cost of the window. That happens rarely enough that I'm still in favor of kids playing ball in the neighborhood.
If heavy drones are crashing down often enough to cause problems (let's say more than a couple dozen times a year in the US), then rules will need to be put in place to curb that, sure.
I think you and I are arguing from different scopes here. My whole premise has been regarding Amazon-owned (or other commercially-owned) drones operating in our skies. You're--rightly--concerned about that guy down the street picking up a 30 lb. machine for fun, and not being responsible with it once it causes damage or injury. I think we can agree in that regulations on the craft and on the operator would apply before it's allowed to fly. Maybe the operator should be bonded and insured, and should have passed some sort of certification. The aircraft itself should meet some minimum standards of safety and redundancy. OR, if the aircraft meets some higher standard of safety and redundancy, then any civilian can operate it because it self-enforces flight safety rules.
Those same rules should also apply to any commercially-operated drones of course, but I've been assuming that, e.g. Amazon, would do this out of self-interest. They don't want to deploy a fleet of drones only to have a massive PR nightmare when things go south.
And after all these rules are in place, in the instances where someone does violate the law and flies illegally, existing insurance and law enforcement schemes would be able to handle it.
All I'm afraid of is cutting off innovation before it can occur, due to an over-abundance of caution. That's all.
I would expect there to be a requirement for an independent parachute system on such drones - independent power, sensing, etc. If it senses that it's falling, it deploys the chute. That module would have to be FAA certified.
There are millions of cars on the road, but you know who to sue if one crashes into the front of your house. License plates and liability insurance would be sensible requirements for drones, too.
This is of course a legit concern, but what should be remembered is that it has to be balanced not against zero risk, but against the risko of the alternative.
Today delivery of goods mostly happen in cars. Cars have an awful lot of accidents and other negative external effects. This is of course all guess-work as long as we don't have real world data from mass drone flights, but I'd assume that drones may come out favorably in terms of accident damage.
> but I'd assume that drones may come out favorably in terms of accident damage
I'm not sure that we should assume that.
Anything that is 100-400 feet off the ground is inherently more dangerous than an equivalent object resting on the ground, simply because of what happens when it fails and gravity takes over.
We can compare automated flying drones to manually-operated cars. We could also compare automated flying drones to automated cars.
The drones scenario might end up safer than now. But I would not assume it.
The trouble with drones, even more than self-driving cars, is that soon, lots of manufacturers will be able to make them. When no-name unaccountable manufacturers make drones for people who buy and fly them anonymously, we won't be able to prevent the havoc that occurs, with laws.
Previously, this was hardly a problem, because very few machines operated autonomously outside of a building. Cars had occupants. Helicopters and planes had occupants. Now, many won't. If a rogue drone with mounted guns starts shooting up a public square, we won't know who launched it unless the manufacturer played by the rules.
That's a serious problem for our society that's coming up. Terrorism has always been a problem of technology, and we're about to make more technology available to the masses, where some fraction of a percent of wackos is going to use it for nefarious purposes. The trouble with unmanned robots which are easily manufactured is that we can't trace who made it, sold it, or launched it. And therefore can't enforce the law on everybody.
I am more concerned about this than governments using the technology.
It doesn't seem all that different from the market for any other regulated product. The FCC approval process for electronics more or less works, not because it's impossible to get around it, but because most people don't go shopping for stuff that's not FCC-approved and most businesses don't want to risk selling it.
Of course there are always black markets, but their scale depends on how much unmet demand there is that can't be satisfied by regular markets.
The difference is that we will have self-directed autonomous technology running rampant outside confined areas, and you only need 0.1% of it to go rogue to cause serious damage.
>If a rogue drone with mounted guns starts shooting up a public square, we won't know who launched it unless the manufacturer played by the rules.
a bomb hidden in a thrash bin or a stolen car parked on the public square. A bullet from sniper rifle from a mile away. No drones necessary. And just imagine what kind of, including chemical or nuclear, havoc can any engineer wreck with tech available today.
>Terrorism has always been a problem of technology.
These things aren't able to follow instructions, target people, etc.
Go ahead and tell me what we will actually do when these things appear.
You can poo-pooh security through obscurity but we're all relying on the "infeasibility" of defeating various systems. Just as one example, reputational attacks followed systematically without rest or hesitation can subvert entire communities and make them not trust each other. You only don't realize what dedicated technology can do to subvert human systems because you haven't had to face it yet.
If every time you launched your startup, an automated process would find, hack in and bring your network down or make a thousand sleeper accounts which will bring it down later, when you couldn't rely on them "just not finding you long enough for you to secure your network", then you'd have a different attitude.
there is a difference between doomsday declaration on every technological advancement and just reasonable basic safety measures.
Specifically for drones - having millions of them in the sky would naturally lead to 2 things - automated traffic control with ground and airborne AESA radar systems that even military would envy (after all an Aegis destroyer have to track only several hundreds objects simultaneously where is drone traffic control over Bay Area would have to track a million) and the second thing - "license plate transponders" similar to car. Any drone without registered transponder will be taken off the sky by a police drone similar like an unregistered car from highway today.
So, yes, somebody would use the drones for crimes like today they can use planes or cars. So what? The drone flight will be recorded in radar and various visual systems. And like in any other case investigation will trace the stuff back to perpetrators. And sometimes not successfully.
And yes, machine guns plus cars did produce the Dillinger gang and the like for some short time. And it resulted in FBI. The way of progress...
So what? So right now when they use planes and cars you can see who was driving it, who made the car, who purchased it.
With drones, you won't be able to see any of that. The criminals who programmed the drones would be free from the law. And not only that, the "police drone" as you say might arrive too late, the drone might already have shot up some people. And then there's always the possibility that despite your best efforts, some drones are gonna fall from the sky on people.
Fair enough, but as the level of abstraction increases in crime it becomes easier for people to commit crimes that wouldn't have otherwise. Remove the human interaction and the guilt that comes along with that situation is suddenly minimized.
I've always seen these Amazon drone delivery stories as an elaborate PR stunt (I'd say an extremely successful one if it is...). But every time the story comes up on here everyone seems to take it very seriously. Am I just being really cynical or do we actually think this is going to happen in the next 5-10 years?
It's not a question of drone deliveries coming in 5-10 years, it's already happening in some places: for example, DHL has a drone in service (albeit as a research project) shuttling urgent pharmacy orders to an island 10 km off the German coast.
The question for the next 5-10 years really if the economics and regulations can change or have changed to make it practical to do this on an ongoing basis for small and larger payloads.
Thanks for that comment. I felt I was the only one thinking that these drones in the sky constitute a major pollution and visual agression. In my case, it triggers the exact same revulsion I have when I discover a big spider walking on my ceiling. Yuck.
At least with trains, roads and airports, we can move away from the source of pollution. With drones, great, nowhere to hide. Pollution comes to you. Not to mention that I am sure soon we will have ads displayed on the side of the said drones. Did I hear "Blade Runner"?
At least with trains, roads and airports, we can move away from the source of pollution. With drones, great, nowhere to hide. Pollution comes to you. Not to mention that I am sure soon we will have ads displayed on the side of the said drones.
For all the negativity in this thread, you've pointed out (in my opinion) the one actual problem with this. It's going to be like living next to a highway in any urban area. Not any noisier than an apartment in new york, I suppose, but disruptive to quieter cities.
I'd guess it simply won't be that big an issue. They'll know where they are all the time, and they'll file theft reports and get police involved. They could add video, too, if they felt the need. In the US, it doesn't seem like organised bands are going to risk criminally attacking autonomous devices to steal parts.
242 comments
[ 5.1 ms ] story [ 481 ms ] threadI wonder what size drones they are envisioning. For example with a typical prosumer drone out there now, it would be completely obliterated by a large helicopter when coming in contact with the moving blades or hitting the windshield or other parts. Likewise the downdraft of the heli would force down anything of that size in it's path.
After all if a drone is such a threat to a heli then what kind of threat is it to a person that it comes into contact with by accident?
THe TL;DR of the episode is - carbon fiber blades directly connected to motors (likely the setup for commercial UAVs) is capable of slicing deeply enough to hit your jugular, others are not (unless the motor is being held solidly and used as a form of saw).
I'm more concerned with UAVs and payloads in excess of 10 lbs falling from 400 feet up onto my head because a wiring connection came loose and cut power to the UAV (happened to me recently, minus the head).
In short, don't fly UAVs around people, or in no-fly zones.
http://blogs.wsj.com/metropolis/2013/09/05/remote-control-he...
"A 19-year-old model helicopter enthusiast was killed Thursday when a toy helicopter he was flying struck him in the head, a law-enforcement official said."
I imagine even the downdraft coming off of a helicopter would be enough to ruin and take out any drones that got within a certain level of it.
Downdraft wouldn't help safety: presumably there are locations above the helicopter from which a drone would be pulled into the rotor.
I see posts sometimes where they write about how great X is and then have a disclaimer of (we're owned by creators of X) as if that was good enough. I just think if you have a conflict of interest, unless the value of the news is significant, writing about X should be avoided altogether especially if it's heavy on speculation, analysis or opinion.
If you're an expert in a certain area, it's quite likely that it's because it's part of your job. There aren't that many people who are both impartial and knowledgeable.
If the story was about American Airlines' employee compensation practices, and it was published by the Air Line Pilots Association, then the same conflict would arise. But if the story is about aviation in general, then the conflict goes away and the pilot is a good source of expert knowledge.
Well, not exactly, just by a company with the same majority owner.
This example is not at all like the situation with Jeff Bezo, Amazon, and The Washington Post. In situations like this, there is the perception (and often the reality) that employees of the media organization would self-censor and not write anything too negative about a company owned by someone that can fire them (or indirectly hinder their careers).
It's a little trickier than that: Not covering something also can be beneficial. For example, what if the Washington Post omitted coverage of bad service at Amazon?
Whats surprising, is that it took something so tech-related to make you notice!
It also has some real issues in that it's glossing over the challenges of steep terrain and obstacles. Most of our air navigation takes place by reference to MSL altitude (i.e. relative to sea level), but this is exclusively based on AGL (i.e. relative to ground level). Not an insurmountable challenge of course but it seems like they are skipping a lot of important parts. Same impression of the "real time data" plan, which sounds extremely complex in the context of aviation.
The overall impression is that nobody has really thought this through particularly, that this is more of a conversation starter at best.
(Relevant: I have a pilot's license)
Having ADS/B or a variant of it for the drones would take care of that issue, as in 2020 every aircraft will require ADS/B.
Anyone touting ADS-B as a realistic solution for avoiding mid-air drone collisions in congested areas is severely naive in my opinion, at least based on current technology. My assumption is that those advocating for it in this context aren't actually familiar with its specs and how it works.
In aviation, separation between flying objects is usually measured in miles horizontally, and a thousand feet or more vertically, and the systems are based on that reality for the most part.
Some consumer drones use a radar altimeter to determine altitude AGL. The hardware is inexpensive. GPS is accurate for altitude in the US using WAAS (extremely small cost to use WAAS over traditional GPS).
> Anyone touting ADS-B as a realistic solution for avoiding mid-air drone collisions in congested areas is severely naive in my opinion, at least based on current technology. My assumption is that those advocating for it in this context aren't actually familiar with its specs and how it works.
Disagree. ADS-B was designed specifically for this circumstance (traffic avoidance).
Disclaimer: I have electrical engineering experience, a private pilot license with an IFR endorsement, and have worked on projects using ADS-B in production environments.
I think the only viable solution would be to have sensors on the drones themselves -- visual and perhaps low-power radar. That way they should be able to avoid other drones and helicopters. It might be necessary to limit operations to VMC, although that shouldn't be much of an issue at below 500ft (you're only really going to be shut down during fog).
ADS-B broadcasts contain a unique ICAO code, 6 hexidecimal characters filled in by the Transponder. The collision avoidance logic is handled by the TCAS unit on aircraft.
While new collision avoidance logic is being developed for unmanned vehicles in order to allow them to legally fly in airspace once the FAA restrictions are changed from see and avoid to sense and avoid, when it is released implementing it on an air frame will not be cheap, or light so will be reserved for larger drones. Think about 100k and 20-30 lbs in equipment for an ACAS-Xu Unit, Transponder, and antenna.
And because ADS-B is an internationally used system, changes like the addition of new messages to provide any other information commercial entities wanted their drone to provide would be unlikely to happen quickly if at all, as they would have to go through committee review.
Many will equip, but nowhere near all, and nowhere near enough to design a system that relies on every bogey having ADS-B out to avoid mid-air collisions.
http://www.faa.gov/nextgen/programs/adsb/faq/#8
I agree. We need to be critical of this idea because, as one of my design professors taught us, designers often "win" because they are the first ones to get their ideas out there and thus control the conversation.
I personally don't want anything dangerous or loud flying within a quarter mile of me. It's a cool hobby to do in a field, but those things in suburban or city environments could be looking into windows to see who's home, aiming bullets at kindergarten teachers or piling up C4 on a mall rooftop with next to no accountability or policing. Cars and people can do all the bad things I'm talking about, but there are license plates and cops. Looking at the frequency of attacks where non-traditional policing is the only option (e.g. online) tells us drone crime will be rampant, perpetrated by a spectrum of interests from foreign governments to financial crime syndicates, abusers and corrupt and paranoid politicians. All other dangerous networks like highways, runways and gas pipelines are tightly regulated. Drones belong in that category.
I guess all the media and government fearmongering since 9/11 has been successful. When someone pitches a piece of tech, people now immediately think of how that tech can be used by Bad People to do Bad Things.
I would expect Hacker News of all places to not succumb to this type of irrational paranoia. But I guess we are all human in the end.
The real issue is that I don't want a drone highway over my house. If I don't want planes flying over my house, all night and day, I don't live by an airport. If I don't want traffic noise, I don't live by a highway. But if Amazon decides that 300 feet above my house is the best route for deliveries or some other logistical exercise, I have far less recourse than I would for a road or an airport.
There is tremendous likelihood that regulating drones will lead to less crime. Banning drones within 1/4 mile of permanent residences and mall rooftops will prevent a large chunk of drone crime.
Banning drones from mall areas is not about "making it more illegal to put C4 there." It's about empowering the law and private citizens to forcibly remove drones from a dangerous areas to prevent drone crime.
If drones are allowed to buzz around malls, there are no police to determine if one of them is piling up C4 on the roof. If drones are not allowed to buzz around malls, as soon as one as spotted in a no-fly zone, it can be taken out to prevent a dangerous crime.
I just looked at the live map for the SF area. A Temporary Flight Restriction over most of SF begins in 39 minutes, surface to 3000 feet, because there is a baseball game at the ballpark.
Autonomous drone operations in urban areas are going to be very tough to control. Drones are going to need their own airborne radar. Maybe only a hundred meters of range, but they're going to need to be able to see details down to power lines. This is available now, but needs more downsizing.[1]
What might work is a system where all manned aircraft have ADS-B (this is coming) and drones also have ADS-B and enough smarts to stay well out of the way of anything bigger. Drones also need short-range radar so they can avoid bumping into obstacles and each other. Think of this as "micro-see-and-be-seen'. It's a lot of electronics to carry, but less electronics than a smartphone.
This is more likely to work than some scheme where everybody phones home to Amazon's cloud.
[1] http://www.autonomoustuff.com/sms-radar-proximity-warning-sy...
As out of the box hobbyist drones can already travel at 50 mph [1] (or more I assume). And 100 meters assuming approach of a stationary object would provide just over 4 seconds for the drone to respond and avoid the object.
While 4 seconds is sufficient time, it will lead to drones flying very close to objects before avoiding them which will not inspire public confidence. Also as faster drones appear that window will become even smaller. They should probably have radar with a range measured in nautical miles to avoid having to implement a new system almost immediately. I would say at least 5 nautical miles, although I feel even that would be on the short sighted end of implementation.
[1]http://www.dji.com/product/inspire-1
The state of the art in radar is that sensing power lines and similar wires is not reliable. 77GHz radars employed in cars can sense people and bicycles, but that's not good enough. There's been much effort put into power line detection for aircraft (helicopters hit power lines about 50 times a year in the US) but no good solution yet. Drones zipping around streets with overhead wires and trees is not yet workable.
Drone Delivery - never going to work. Lots of solar power - too expensive. Autonomous cars - too risky
This is supposed to be a community of "Hackers" making cool stuff, and here we have a company talking about making cool stuff, and every single comment, without exception, is negative.
Yes, this is a very difficult problem to solve. Yes, there are many things they have not addressed yet. Yes, this needs a lot of work. No, they shouldn't just start doing this and see how many crashes happen. But yes, they need to keep talking, testing, learning and working towards a future where this is possible.
Does anyone in their right mind honestly think that in less than 150 years there will still be a single human-controlled vehicle (plane or automobile)?
Of course it's coming, and these guys are starting it. Now.
The same negativity could be said of the Wright brothers when they told people they were going to fly, or Henry Ford when he set out to mass-produce the automobile, or the people that said we'd have a global computer network and a computer in every home.
I honestly don't understand why the community here isn't more open to new ideas. Even ideas that never come to fruition are still worth analyzing. There is always something to be learned.
We should be talking about the interesting parts of this challenge and how they could be overcome, not saying "Not safe enough" Every. Single. Time.
</rant>
You do realize that we don't even have clean water on a global scale?
You: You do realize that we don't even have clean water on a global scale?
I think the parent commenter's point was that it's pretty naive to think that there will be no more human-driven vehicles at all. Change will happen quickly in some parts of the world, and other parts will take a long, long time to catch up.
The technology to filter and clean water has existed since the Roman era, but its implementation is highly dependent on political and economic factors (fun fact: western civilization didn't match the Romans' quality of water engineering until the mid-19th century).
The technology for automated cars could well exist in 150 years, and its implementation depends on the cycles of industrial production and consumption. It's likely that much of the western world will have automated transit in 150 years, because everyone will eventually get rid of their 2015 toyota for a model 2077 automated Tesla-google car.
With automated transport, the early movers could absolutely dominate the future of transport.
or may be a community of voyeurs who like to watch the Hackers making cool stuff?
And it never ceases to amaze me how the tech community so easily blows off legitimate safety concerns.
You can't just "reboot everything" after a drone wipes out a Cessna or smashes a person's skull. Any drone capable of delivery is going to be big, powerful and not light.
Everybody thinks of drones as "cute little things" when a delivery drone is going to be more like a flying lawnmower like this: https://www.youtube.com/watch?v=l5FqYiZb_5s
I seriously doubt any drone proponent will be dismissive of safety concerns. If there are people who don't care about the safety concerns drones bring, this is just the other side of the same coin with people so resistant to the ideas of autonomous vehicles. Both are people not forward looking in their own way.
When Amazon says, "hey we want to use drones," reactions shouldn't be, "hell no there's no way you can make that safe," and instead, "great idea, let's make sure there is a discussion at large to reach a consensus on the safety standards required to make that happen."
Most engineers are conservative and skeptical. If anything, as someone who is an aerospace engineer by education, I'm always surprised by the wild eyed optimism around here.
> The same negativity could be said of the Wright brothers when they told people they were going to fly
It could also have been said of the generations of idiots that jumped off cliffs to their deaths in flying machines that had no hope of working.
In any case, I don't think people are pessimistic about the technology. They're pessimistic about the ability of our social institutions to accommodate and manage the effects of that technology.
That's because the HN community has a high bar, and most "new ideas" are bullshit (when it's not a rehash of failed old ideas shot down 100,000s of times, like "perpetual motion machines").
Partly I think this is people setting themselves up to be able to say "I told you so!" at a later time. You look less silly predicting a failure that doesn't happen than if you extol the virtues of an idea that later turns out to be complete hogswash under further scrutiny.
Though to put a more positive spin on it: critical thinking is a much neglected skill. Sometimes things genuinely aren't well thought out, and identifying and documenting a problem are the first steps to someone coming up with a solution.
https://en.wikipedia.org/wiki/List_of_tallest_buildings_in_S...
Food for thought
That much is covered in the diagram in the article: anything near an airport is designated no-fly for the drones.
Avoid roads and sidewalks.
> If they're going to have a non-zero number of these drop from the sky, they better know they're not getting that drone back, for starters.
While you may think property rights don't exist solely because of the word "drone", you'd be mistaken.
I have serious doubts you're going to see significant drone traffic in the area where you live (based on your profile info).
Of course they will. Why wouldn't they? You have their drone and a photo of a gaping hole in your roof.
> If they're going to have a non-zero number of these drop from the sky, they better know they're not getting that drone back, for starters.
Well, they'll probably make damage reimbursement contingent upon your return the drone.
(If the law doesn't compel you, Amazon might pay extra for return.)
1) http://nypost.com/2015/06/09/new-york-post-editor-and-film-c...
But in terms of things falling out of the sky, it will definitelly be interesting (the Chinese curse kind of 'interesting').
Obviously supersonic booms are a different animal, but they had a roll in keeping the tech from going mainstream here in the US:
https://en.wikipedia.org/wiki/Oklahoma_City_sonic_boom_tests
I feel like a delivery drone is the absolute most complicated way to solve this problem other than somehow involving rockets.
Drone limitations: - Needs great weather. - Package size and weight exceptionally confined. - Exceptionally limited range. - How to actually get package from drone to human? Drop? - Night flight? Yes / no? Probably not.
Car limitations: - Traffic.
Where I live it is not an issue, but it is theoretically possible that there might be a location on the planet right now where Fedex trucks crash into UPS trucks, and Dominos pizza cars are running Chinese delivery cars off the road, on a daily basis. However that weird situation existing would still be a workable system for perhaps 99.9% of the worlds surface.
There are also practical scaling issues. The local pizza delivery joint cannot produce fast enough to saturate a 2-D delivery system during the superbowl, just can't pump out pies fast enough. Naturally going to 3-D delivery with maybe 100x the delivery capacity because of 3-D isn't going to alter their delivery rate, the skies will be pretty empty. Even if you drive there to pick up, there will be a 2-hr wait for pizza. Drones and trucks are expensive and the finance guys would prefer the absolute minimum be owned and deliveries be made 24x7. That would imply we're not going to get drone congestion until we have deliveries at 3am. I'm just not seeing people order enough raw "stuff" to keep delivery services busy enough.
I'm skeptical on drones, and I think all this is an effort by Amazon to get in the news. But this doesn't seem like the problem.
Amazon's idea is to have airspace dedicated to autonomous drones, who can go from a to b in a straight line and mostly just have to worry about not hitting other drones (or buildings or birds or what have you). You can do that with some decent sensors and some simple physics simulations.
That's not to say that airspace is totally unregulated, but the regulations tend to be much broader (AFAIK; I'm not a pilot).
It is only the severity of an accident that increases, not the likeliness.
There are dozens of reasons why an autonomous drone is greater than an autonomous car, use your imagination.
Not for long, apparently.
Apart from that, it still isn't obvious to me how this is better than an electric delivery truck. Cars are much more energy efficient than drones (especially helicopter-like ones), doubly so if you load the car to its capacity. Note that in case of drones, the mass of the package is only a small percentage of the overall mass of the vehicle, so most of the energy used will scale up with number of drone deliveries. I'm pretty sure taken altogether cars will win in terms of energy use.
But even if they find a way to handle the traffic, I was just saying that there will be traffic and that this poses serious problems.
How will they put deliveries into letterboxes? Or will there be a special Drone Nest Drop Off Utility Safebox[tm]?
And so on. The more you think about the idea, the less practical it gets.
I'm sure the problems can be solved. Obviously, you could retrofit your doorbell to include a drone code. Or make an app (it's always an app...) that alerts you when the drone is near.
This is perfectly fine, unless you don't want to change your doorbell and/or don't have the right kind of phone and/or don't want your stuff stolen.
There's a whiff of "Reinventing Yesterday's Technology for Tomorrow" about the idea. It looks a lot like a high school science show-and-tell trying to turn into a delivery system that can compete with 75 years of professional logistics expertise.
I don't usually do "not invented here", but there are so many very obvious things that can go wrong, and so may ways users need to change what they do, that I think debates about air traffic control are a long way down the list of show-stopper issues that need (literally) bullet-proof solutions.
I would like to see a startup for what amounts to background checked, bonded, doorman service. Someone in my neighborhood I could trust to take deliveries, stuff like that. Outsourcing trust, I guess.
I could see drone delivery company "tipping" some old retired guy to solve their "last tenth of a mile" problem as he walks down the block to my house.
Alternately make all uber (or insert other) driver watch some training videos and tell them to be at xyz address at abc time and they verify safety and walk the last 50 feet and hand deliver, for a modest tip of course.
Presumably if I watched the training videos and didn't have badminton nets and kids pool and ham radio antennae in my backyard the drone could deliver directly to me and save me the cost.
For one, you could always have a human operator coordinate the landing and delivery process. They could coordinate delivery using a phone call.
Secondly, you could have a drone deliver goods to a distribution truck (possibly with a special landing-pad marker on top,) and a human could then deliver the package manually.
No advanced technology needed for either of these, it's just communication + drones.
You have an automated road vehicle that acts as an "aircraft carrier" for drones. Something roughly the same size as a delivery truck today. It gets loaded up at the distribution center with packages and a few drones.
It picks a route that gets roughly near all of the deliveray locations. As it goes by, a drone is dispatched to take the package a couple of blocks to the person's doorstep and then returns to the "aircraft carrier".
When the drone gets back to the truck, it plugs in an recharges.
This uses drones for what they're best for: quickly and efficiently navigating small complex spaces like side streets and front yards when there is only a small payload you need to deliver. The truck addresses the shortcomings: limited flight time and payload capacity.
Canada, specifically British Columbia, is hardly "overseas".
http://www.thecanadianpress.com/english/online/OnlineFullSto...
Even though I am quite fascinated by the innovation of this idea, not so much about the practicality of it. I believe it is good for rural areas where highway system is not streamlined and major shipping carriers don't encourage frequent routes.
http://www.slate.com/articles/news_and_politics/explainer/20...
Has anyone done an analysis of the crash rate for large drones and likely casualty rate on the ground?
I'm waiting for the ability to transmit "bird alerts" to chase away the drones I don't like.
I have no idea if it will offset or not, but I could see—even with a few thousand drones crashing every year—that a net benefit could still be realized because people are not driving to the store for some of their shopping.
But if drones are replacing car trips to the store, then there's a good opportunity there to offset automobile fatalities
In Houston there are only a couple of logistic centers for UPS and FedEx and whatnot. You wouldn't have a drone fly 40 (or maybe 100) miles round-trip to deliver a package all the way across town.
But if every regular delivery truck has 2x or 3x drones on top they can make small package deliveries without the driver stopping on his route as he/she delivers the bigger packages. The ones too heavy for a drone to lift.
The drones could make deliveries during the driver's lunch break which is a nice bonus. An hour with the truck parked in the same spot.
You could also have drone-specific semis that go on a very slow rotation around the city, moving a few miles quickly and then stopping so the drones can make deliveries from a "home base".
Drones are a good last mile solution but will never replace the back haul network. Wheel on pavement has incredible efficiency versus rotor through air.
How this will exist in the future in practice is pure speculation at this time. There are many different factors (autonomous vehicles, smarter routing techniques, routing changes as the burden is shifted to non-truck delivery systems, and increased demand for small package delivery as more shopping is done online, for example) that could dramatically reshape the delivery landscape.
Furthermore, heavy packages delivered to consumers in the US, like those containing appliances and furniture, are already routinely delivered by in-house or smaller delivery services rather than UPS, FedEx, or US Mail.
It would be interesting to see data on what % of amazon purchases are above the payload of a drone - i imagine they crunched the numbers themselves and determined it was low enough to justify continuing to push this issue (even after the FAA tried to ban[1] drone deliveries altogether)
Theyll look at $/package and time/package for drones and delivery drivers.
I imagine once you account for driver salary and benefits, cost of vehicles, cost of fuel and maintenance, cost of insurance for drivers and vehicles, cost of management structure to maintain drivers, schedules, routes, etc - the $/package for drones likely make up for any disadvantage they currently have in time/package
Also, drones can deliver packages 24/7/365, no reason you cant wake up to a package delivered overnight (maybe even your morning kindle (tm) e-ink newspaper)
[1]http://techcrunch.com/2014/06/24/faa-bans-the-use-of-drones-...
The fallacy in that thinking is left as an exercise for the reader.
Also, we have no idea what the drone-aircraft collision risk is. That seems like something that could be easily solved by strict control over who flies where in the sky. I'd guess that more people will die from these things falling on their head.
The DJI Phantom for example only has 4 rotors and a single battery, so any one failure of a component in the power train or controller will likely bring down the entire device.
The type of drone Amazon is likely to deploy widely in this scenario would of course be much larger, but would also have many more rotors, think of a copter with 8 arms and 2 motors per arm, with the power trains wired up as 4 independent systems each with 4 opposing rotors and it's own battery capable of keeping the device flying independently. Flight controllers can also be built hilariously cheap, so having 3 or 5 in a voting configuration is equally simple. Now you have a machine that can stay airborne and under control with over half of the rotors disabled.
When drones are making millions of flights a year, even a one in a million malfunction is likely to occur many times a year.
It's a shame we have become so risk adverse as there are plenty of things way more likely to kill you then drones.
Me: "No."
Police: "Did it cause more than £2000 worth of damage?"
Me: "I don't think so."
Police: "Call your insurer. This is not a police matter. Bye."
Insurers will presumably say "that's not covered" up until the point where they have an extra charge for drone insurance when they can say "you don't pay for that cover".
Where I live, if a drone crashed in flames, it would start a fire that would replay the Oakland Hills fire [1] with people getting killed. If a car were to crash on Sand Hill Road, not so much.
[1] https://en.wikipedia.org/wiki/Oakland_firestorm_of_1991
- Cars travel defined roadways having mandated brush clearance zones that diminish the likelihood of a vehicle crash/fire propagating into uncleared country. A fiery crash is very likely to be noticed and reported immediately, with no uncertainty about its location. Fire responders will be quickly dispatched to the correct location. The road on which the crash occurred will very likely give responders good access for their suppression work. If evacuation is needed, it's easier to direct evacuees to routes that will be most effective in getting them out of danger.
- A drone isn't constrained to defined roads. Compared to the car crash scenario, when a drone crashes away from roads and ignites vegetation, more time will pass before the fire is noticed - reported - located - and fire suppression resources put on site. Away from roads, this means airborne fire suppression and that adds at least 30 minutes to the response loop, even when (as now) there are airborne assets on alert. So an off-road ignition is almost guaranteed to have a longer incident response time than a car crash. Same is true of other off-road fire events like careless hiker cigarette toss, camping fires, etc. Directing evacuees which way to go becomes more difficult.
- Favorable firestorm circumstances: even small fires can combine with topography to generate fast up-slope winds that, while local, are very effective in broadcasting burning material that ignites locations remote from the original burn.
Summarizing, if a drone crash starts an initially unobserved fire in an uncertain location that's tough to access, there's a larger chance that it will become a danger to the local rural population than one started by a car crash.
I'd like to think that the cost of a passenger aircraft isn't the only reason why they come under more safety and regulatory scrutiny than a motor car. I'd hope that it's because they have an additional severe failure mode that cars don't called falling out of the sky.
I realise that I might be going on about this and I'm a layman, but it seems a bit overlooked that any object that is airborne and needs to be made safe surely starts at a large handicap compared to objects on the ground.
http://www.houselogic.com/home-advice/home-insurance/homeown...
"... either a major claim or too many claims in a given period, no matter how legitimate, could lead to losing your house’s coverage altogether..."
Sure, if it's a known drone from a large company, I'm sure they will have insurance, but what do you do when you have a $5000 hole in your roof and Amazon claims that the "Amazon" drone that you found is not theirs - someone slapped an Amazon label on their own drone to get away from the insurance requirements.
But as drones become more ubiquitous and the big ones become cheaper and more useful, then the chance of damage increases.
Amazon isn't just talking about legitimizing their drones, but legitimizing drones in general with an "anything goes" airspace below 200 ft.
I wouldn't make a homeowners insurance claim for a baseball through the window since claims mean higher rates and multiple claims can lead to blacklisting among multiple companies. But a 20 or 30 lb drone dropping 500 ft to my roof may cause thousands of dollars of damage, and unless I can identify the drone's owner (or even if I do and he has no money), then either I or my insurance company is on the hook for repairs.
If heavy drones are crashing down often enough to cause problems (let's say more than a couple dozen times a year in the US), then rules will need to be put in place to curb that, sure.
I think you and I are arguing from different scopes here. My whole premise has been regarding Amazon-owned (or other commercially-owned) drones operating in our skies. You're--rightly--concerned about that guy down the street picking up a 30 lb. machine for fun, and not being responsible with it once it causes damage or injury. I think we can agree in that regulations on the craft and on the operator would apply before it's allowed to fly. Maybe the operator should be bonded and insured, and should have passed some sort of certification. The aircraft itself should meet some minimum standards of safety and redundancy. OR, if the aircraft meets some higher standard of safety and redundancy, then any civilian can operate it because it self-enforces flight safety rules.
Those same rules should also apply to any commercially-operated drones of course, but I've been assuming that, e.g. Amazon, would do this out of self-interest. They don't want to deploy a fleet of drones only to have a massive PR nightmare when things go south.
And after all these rules are in place, in the instances where someone does violate the law and flies illegally, existing insurance and law enforcement schemes would be able to handle it.
All I'm afraid of is cutting off innovation before it can occur, due to an over-abundance of caution. That's all.
I'm not sure that we should assume that.
Anything that is 100-400 feet off the ground is inherently more dangerous than an equivalent object resting on the ground, simply because of what happens when it fails and gravity takes over.
We can compare automated flying drones to manually-operated cars. We could also compare automated flying drones to automated cars.
The drones scenario might end up safer than now. But I would not assume it.
Previously, this was hardly a problem, because very few machines operated autonomously outside of a building. Cars had occupants. Helicopters and planes had occupants. Now, many won't. If a rogue drone with mounted guns starts shooting up a public square, we won't know who launched it unless the manufacturer played by the rules.
That's a serious problem for our society that's coming up. Terrorism has always been a problem of technology, and we're about to make more technology available to the masses, where some fraction of a percent of wackos is going to use it for nefarious purposes. The trouble with unmanned robots which are easily manufactured is that we can't trace who made it, sold it, or launched it. And therefore can't enforce the law on everybody.
I am more concerned about this than governments using the technology.
Of course there are always black markets, but their scale depends on how much unmet demand there is that can't be satisfied by regular markets.
a bomb hidden in a thrash bin or a stolen car parked on the public square. A bullet from sniper rifle from a mile away. No drones necessary. And just imagine what kind of, including chemical or nuclear, havoc can any engineer wreck with tech available today.
>Terrorism has always been a problem of technology.
centuries ago Church tried to outlaw crossbows:
http://militaryhistorynow.com/2012/05/23/the-crossbow-a-medi...
Safety through obscurity works only for very cutting edge technologies and only for the very short time.
Go ahead and tell me what we will actually do when these things appear.
You can poo-pooh security through obscurity but we're all relying on the "infeasibility" of defeating various systems. Just as one example, reputational attacks followed systematically without rest or hesitation can subvert entire communities and make them not trust each other. You only don't realize what dedicated technology can do to subvert human systems because you haven't had to face it yet.
If every time you launched your startup, an automated process would find, hack in and bring your network down or make a thousand sleeper accounts which will bring it down later, when you couldn't rely on them "just not finding you long enough for you to secure your network", then you'd have a different attitude.
Specifically for drones - having millions of them in the sky would naturally lead to 2 things - automated traffic control with ground and airborne AESA radar systems that even military would envy (after all an Aegis destroyer have to track only several hundreds objects simultaneously where is drone traffic control over Bay Area would have to track a million) and the second thing - "license plate transponders" similar to car. Any drone without registered transponder will be taken off the sky by a police drone similar like an unregistered car from highway today.
So, yes, somebody would use the drones for crimes like today they can use planes or cars. So what? The drone flight will be recorded in radar and various visual systems. And like in any other case investigation will trace the stuff back to perpetrators. And sometimes not successfully.
And yes, machine guns plus cars did produce the Dillinger gang and the like for some short time. And it resulted in FBI. The way of progress...
With drones, you won't be able to see any of that. The criminals who programmed the drones would be free from the law. And not only that, the "police drone" as you say might arrive too late, the drone might already have shot up some people. And then there's always the possibility that despite your best efforts, some drones are gonna fall from the sky on people.
http://www.theguardian.com/technology/2014/sep/25/german-dhl...
http://www.dhl.com/en/press/releases/releases_2014/group/dhl...
The question for the next 5-10 years really if the economics and regulations can change or have changed to make it practical to do this on an ongoing basis for small and larger payloads.
At least with trains, roads and airports, we can move away from the source of pollution. With drones, great, nowhere to hide. Pollution comes to you. Not to mention that I am sure soon we will have ads displayed on the side of the said drones. Did I hear "Blade Runner"?
So, just like cars?
Godspeed, brother.
Korean-Americanly yours, F
Even on an aesthetic level, it will be amazing to look at the city sky and see hundreds of lights buzzing about, blade-runner style.
Could be a lot of noise pollution though.