Product That Will Eventually Be Made: Getaway Drones
patrickjamesmcguire.com
patrickjamesmcguire.com
First, the drone must be outside to be able to escape quickly. That's a problem, as consumer drones aren't designed to spend weeks sitting outdoors. Even if a drone was weatherproof, it would need its batteries topped-up from time to time.
Second, you will not have access to your phone or other signaling devices if you are getting busted. It would be very hard to trigger the drone surreptitiously.
Last, police are going to notice a drone flying away during a bust. And they won't be happy about it. For all they know, the drone could be carrying a bomb. This escalates the situation and makes a bad outcome much more likely.
If you want a better use of drones, here's one: First response. Place drone "nests" every 500 meters in urban areas. Cell towers might be convenient locations. When ShotSpotter[1] detects gunfire, dispatch the closest drone to the scene. It can arrive within 20 seconds, record footage, and relay it to police. Currently, police and medics wait for backup before going to the scene of a shooting. But with drones to send the all-clear, they could treat victims sooner, saving lives.
Future upgrades could even allow tracking of suspects. This would make it much harder to get away with firing a gun in a city.
Is there some reason you wouldn't use stationary cameras to do this ??? I mean... it seems like a lot of extra expense and technical hurdles for not much more payoff. Actually... I'm pretty sure there isn't ANY payoff as far as benefits beyond those conferred by the use of stationary cameras goes.
And the technical hurdles are pretty large. You're in an urban area... so how do you get good lines of sight ??? You have to fly around quite a bit... and you'd almost certainly need human controllers, increasing the expense. Drones could crash during incident response, and in an urban area, ie-densely populated, that definitely would entail some collateral damage. Potentially even severe human casualties. Hackers can potentially boost your drones and then they "own" them. I mean... the list goes on and on.
I think that the "better" use you gave is actually one of the WORST ideas you could have for drone use. I mean... we have cameras right now ??? Why would we not just set up a monitoring system... if we wanted a monitoring system ???
So where's the problem? Have two officers on duty, you don't have gunshots that often to justify one officer per drone 24/7.
> I mean... we have cameras right now ??? Why would we not just set up a monitoring system... if we wanted a monitoring system ???
You could do this in the UK, but not anywhere else. NCIS LA ain't real.
Yes, some places might have poor sight-lines. That's a great example of where drones beat static cameras. Places with poor sight-lines would require many static cameras, and they'd need to be updated to reflect new construction. Drones can fly around obstructions, making them more effective than cameras in these situations.
Human controllers would be useful, but cost wouldn't be a problem. You don't need a deputized police officer on "drone duty" at every station. More likely, controllers would be contractors servicing a large area. Software would dispatch the drone to the location of gunfire and a pilot would explore the area to give the all-clear.
"Severe human casualties" are impossible. These drones would weigh less than 10 lbs and carry no weapons. They'd be no more dangerous than personal drones today. Concerns about "hacking" are a movie plot threat. The risk of hacking is the same for any application of drones, including home delivery.
I think I've addressed most of your points. Still, it's annoying that cleaning up a mess takes so much longer than making one. I think your comments could be much higher quality if you made a few thoughtful objections instead of replying with every objection that comes to mind.
Persistent Surveillance Systems does this, but not as a response system. They just photograph the entire city from the air once per second and when a shooting happens they rewind, then trace where the suspect came from and went[1]. They don't use drones though.
[1] http://www.washingtonpost.com/business/technology/new-survei...
I'm basically calling bullshit on the extreme formulation of drone delivery. There will be some niches that they colonize, and a few that they monopolize (they're going to be a boon for postal/small package deliveries to farms/exurban housing -- Post Office self-driving van goes past without stopping while drone shuttles packages to doorsteps a quarter of a kilometer away) but plenty of areas where they're going to be forbidden due to the unacceptable risk of them falling on someone's head.
Here's maybe a better use:
Put weatherproofed power sockets with fiber connections on rooftops every 50 meters in urban areas. Then hang your 5G picocells from drones, and dynamically re-structure your mesh of cellphone bases as demand shifts through the day (by physically lifting and transferring the base station to a new fiber/power feed as necessary). Goal being to feed a maximum number of radio channels where demand is high, while avoiding black spots.
It's true that I can't get Instacart of some of the other same-day delivery services found in urban areas but then there's not enough density of the right kind of customers to support a Whole Foods store either. (I can get Peapod though I don't much care for it.)
All of which is a long way of saying that I don't really see the non-niche problem that even an idealized drone delivery system aims to solve.
>Post Office self-driving van goes past without stopping while drone shuttles packages to doorsteps a quarter of a kilometer away)
Even this is largely a solved problem. Private roads in rural areas usually have some sort of lockbox at the end of the road.
Part of the idea is that you'll get same day delivery for a reasonable price. In fact, it could be cheaper than normal stores (although probably not walmart) since it'll just be a warehouse with a drone fleet.
The horrors of peace appear to be appalling than the horrors of war. The automobile looms up as a far more destructive piece of mechanism than the machine gun. The reckless motorist deals more death the artilleryman. The man in streets seems less safe than the man in the trench. The greatest single lethal factor is the automobile. It left shambles in its wake as it coursed through 1923.
We know how that ended up.
(Self-driving cars can't come soon enough; 90% of RTAs are the result of human-driver error ...)
Much of the public viewed the car as a death machine. One newspaper cartoon even compared the car to Moloch, the god to whom the Ammonites supposedly sacrificed their children.(...) Pedestrian deaths were considered public tragedies. Cities held parades and built monuments in memory of children who had been struck and killed by cars. Mothers of children killed in the streets were given a special white star to honor their loss.
And yet, cars still went to dominate cities and landscapes. I think this tells us something about how drones might become accepted despite early opposition.
The point is there are solvable problems that can make drones viable. I've done a great many long distance autonomous drone flights (we're talking miles) to then have it land within 3ft of where it was supposed to. The tech is very close to being up to the task (3-5 years) the regulation however is not.
Food for thought Cars kill 40k people per year in the US, even if a few are killed by drones the number of deaths avoided by having less car trips would likely well more then offset those killed by falling drones.
One more thing, if cars were invented today they would most likely be made illegal due to the potential for death (40k/year). Aside from a handful of freak accidents drones don't kill people yet there are a great many people trying to ban them out the the fear that they might. Now compare that to the 40k number for cars.
Same could be said about airplanes 100 years ago, but now I have planes flying over my head about 20 times a day. Even in the post-9/11 world.
The safety of drones will only increase, and with such a great financial incentive as the whole delivery industry, it will do so rapidly.
I hate to be a negative Nancy about this, but [citation needed]. I fly a multi-rotor coptor now, and when the battery hits a certain point it just drops out of the sky, wherever it might happen to be. This is just a toy model, so it doesn't have some of the fancier electronics a delivery drone might - but there's nothing inherently stable about a multirotor like there is with a plane.
Even the best motor failure "recovery" I've seen so far is still a multirotor going into a barely-controlled death spiral - something which would still cause a lot of damage to whatever it hit. Imagine it for a moment: a 20-30 pound object falling in an unpredictable (to humans) spiral at speeds faster than you can run, with between three and seven 10 inch props moving at near supersonic speeds at their tips.
I love 'em, but they're dangerous machines with lots of fast-moving parts, and need to be treated as such.
The other is what DJI is already exploring with their "DropSafe", ie., parachute. A simple software tweak (when accelerometer detects falling, shut down motors and deploy parachute) could make most falls reasonably safe (and less costly to the owner).
With cameras, they could also try to avoid flying over humans.
Finally, in cities, they could fly most of their way over the roofs of building, rendering a fall unlikely to hit anyone.
still over-engineered. look how it's done: https://youtu.be/Wp2CEJb4oyY?t=71
Helicopters and planes are maintained to a ridiculous degree. Without getting that kind of safety into drones, they will fall down on people's heads left and right. Especially once the drone traffic gets dense enough that they can run into each other.
"Drones" may be used for drug smuggling, though.
You mean already are:
http://www.cnn.com/2015/01/22/world/drug-drone-crashes-us-me...
1. A couple of kilograms of drone, that typically won't kill you if it runs into you.
2. A couple of thousand kilograms of vehicle, heavy enough and moving fast enough to kill or cause horrific injuries in even a glancing collision, moving around not up in the air where there is open space but at street level where people are trying to walk around, the vehicle itself being completely brainless and relying entirely at every moment on the control of a fallible, easily distracted human operator, and to cap it all, most likely running on burning fossil fuel.
I put it to you that the first option has far better safety properties than the second. If you're going to propose they both be banned, okay, at least that's logically consistent. The one thing you can't do with any sort of consistency or moral authority is criticize the first option while giving the second a free pass.
A vehicle operated by human beings can deliver a lot more packages a lot faster, while their accident rates aren't nearly as sky-high.
There is a very obvious difference between trucks and aircraft: the third dimension. Trucks operate on streets and roads, which are clearly identified and known to pedestrians as a hazardous location where extra attention to the rules and conditions is required. While streets and roads could be made safer for pedestrians (and, often, should!), there are literally billions of pedestrian-vehicle interactions every day that are perfectly safe. This is possible because a street is largely a 2- or even 1-dimensional environment with clear, well-known rules for sharing safely and (generally) good visibility so that every participant can get a full mental picture of that environment.
Aircraft operate everywhere, and at an altitude well above where most people would think to look. The danger from aircraft is not that a 10kg craft rams you horizontally at 5 m/s while it flies 1 meter above the same sidewalk you're using (though even that would probably cause significant injuries). It's that a 10kg aircraft you never saw, thought to avoid, or even had an opportunity to avoid falls 100 meters onto your head. That, by the way, is going to "kill or cause horrific injuries" every single time.
It's fine to hate automobiles. They're not perfect. They do have collisions, and sometimes those collisions kill humans. That's unfortunate, and worth improving. Most of them also burn fossil fuels, which needs to stop (in most of the world, drones will either burn fossil fuels or have batteries charged by electricity generated by a coal-fired power plant, so they're not better). But your "consistent solution" is absurd: a near-total elimination of economic activity in exchange for a very modest reduction in the human death rate. Which reduction would be immediately reversed and then some by the inability of hospitals to procure medical supplies, among many, many other problems.
You should probably think about learning to accept a need for nuance. It really helps in thinking about solving problems, which is what engineering is all about. Insisting that everyone must either agree to ban all automobiles or accept the uncontrollable and unavoidable risk of having others' property fall on their head anywhere at any time is not a constructive position and will not help you solve problems.
A more useful approach might be to ask yourself what would be required to make UAV flights safer. Are there environments or applications for which these problems could be minimized? What regulations or restrictions would help? Or maybe you'd like to tackle pedestrian safety in urban areas, a much better-studied set of problems. Or maybe you'd like to do some research, gathering some data on UAV failure rates and modes so that you could inform regulators as to which risks matter most.
That is, if you want to do something useful. That wasn't clear from your post.
'Drones could in some circumstances be dangerous' implying 'we should look at how to make them safer' - yes, I'm all in favour of that.
Many police forces use Glock 17 pistols, which have a standard capacity of 17 bullets[0]. Ignoring how fake the muzzle blast looks, the "officer" empties his 20 round clip into the air. I can't find a citation, but I believe police are not meant to fire unless they are in danger. While an officer could certainly ignore this, one would know that they could have disciplinary action take for emptying an entire clip haphazardly into the air in a busy location.
maybe multiple balloons will even level out, so they don't all burst in thin air but stay somewhat low?