Slashdot Mirror
An Air Traffic Control System For Drones
An anonymous reader writes: Personal drones are become more popular, and many companies are trying to figure out ways to incorporate them into their business. So what do we do in 10 years, when the skies are full of small, autonomous vehicles? NASA and a startup called Airware are working on a solution: air traffic control for drones. "The first prototype to be developed under NASA's project will be an Internet-based system. Drone operators will file flight plans for approval. The system will use what it knows about other drone flights, weather forecasts, and physical obstacles such as radio masts to give the go-ahead. Later phases of the project will build more sophisticated systems that can actively manage drone traffic by sending out commands to drones in flight. That could mean directing them to spread out when craft from multiple operators are flying in the same area, or taking action when something goes wrong, such as a drone losing contact with its operator, says Jonathan Downey, CEO of Airware. If a drone strayed out of its approved area, for example, the system might automatically send a command that made it return to its assigned area, or land immediately."
Drone or RPV?
Passionately Indifferent
This is a top to bottom, centralized way of handling the problem.
Obviously a decentralized way of organizing the flight would work a lot better, but it would require coding some protocols on the drones and governments want to keep control.
This is a stupid solution, having to file a flight path befor the flight and waiting for the government official to give you the green light.
When his defense asked, "Which computer has Jon Johansen trespassed upon?" the answer was: "His own."
Wouldn't this just be... skynet?
1) a protocol
2) sufficiently wide approval of said protocol
So how do we get that ? In our world, there are five ways to build systems: a) technoloy-driven ( done by the Gyro Gearlooses of our world ) 2) purpose-driven systems ( MS Office et al., ain't gonna work here ) 3) sociotechnical systems ( may work here ? ) 4) politicotechnical systems ( basically, things like the entire Internet, or national highway networks ) 5) open source systems ( seems to be the best candidate here ? )
Religous speak to God. Insane are spoken to by God. When all shut up, one can finally hear Shostakovich in peace
I build and fly these "drones" (multirotors) for fun and work, the type of hardware that would be required wouldn't fit in most hobby level "drones" that are seen causing all the problems (DJI Phantom's etc) and the infrastructure to provide what the above implies would be extensive.
For commercial UAV's of size sure no problem, larger, more powerful radio systems and greater payloads.
At best, with the current people in charge, I can see what I do becoming illegal, or too expensive to participate in.
It's a shame really, because of idiots flying over people and houses, doing "altitude test" through clouds and wondering why their craft dies in mid air, to the small number of pervs actually looking in someones window it comes down to a simple statement.
"Can't have nice things"
I've also learned a bit about human psychology and the intelligence of the average person when they see a multirotor and react with fear.
One woman going so far as to say "Where can I go you won't film me" having walked through a parking lot with security cameras pointed at her and all whilst txting away on her iPhone and through a park full of people with their phones (and one even had Glass on) to where we were flying.
Yes this helicopter is the problem.... (sarcasm) Cow goes moooooo
And people wonder why I'm a misanthrope.
"If any question why we died, Tell them because our fathers lied."
The biggest concerns are keeping them away from manned aircraft and limiting how many are above crowds. I know the ones in use today are small, but you still don't want idiots buzzing the crowd at sporting events, etc.
My apologies if it seems I'm duplicating the post "Name" saying "Drone or RPV?". These things are not autonomous drones; they are actively controlled by people. There is no ATC of the things in the air; it's all about the various people wherever they happen to be on the ground.
There's a park near us where people fly RC planes. Fun to watch, and people keep them over the park, and there's no question they're controlled. The first time someone put up a multi-rotor, though, someone asked, "Is that a drone? Can it go by itself?" No. It's an RC plane just like everything else. And if you keep it over the open land in the park, and stay away from people's windows, you'll be fine.
That is a blatant case of techno-optimism, and a surefire recipe for failure.
Religous speak to God. Insane are spoken to by God. When all shut up, one can finally hear Shostakovich in peace
What is the best load for a drone? I'm thinking #4 buck shot.
This sounds like a centralized style solution. It only works if communication between any drone and the central server/agency is maintained. For some parts, like flight planning, this is fine. But for collision avoidance I don't think this will cut it.
In the shipping world we already have a decentralized system called Automatic Identification System (AIS). Every vessel broadcasts its position and course on a common radio channel. Other vessels listen and if equipped with collision avoidance systems can take evasive action. Something similar for Drones could be imposed by the FCC, like it is on ships by the International Maritime Organisation.
This would suffer from the same drawbacks (ships can fake their identification), everybody can listen to broadcasts, but it would help solve 95% of the problem.
A single helicopter already turns heads and if it hovers for too long, people will complain about the noise. How do you expect this to work when hundreds of drones are buzzing over our heads every day? People will get very aggressive and drones will be downed in any way possible.
Until there's a silent anti-gravity system (McFly, are you listening?), general use of drones will not take off..
To Terminate, or not to Terminate, that's the question - SCSIROB
It seems to me the real problem is separating drone traffic--including drones bought by amateurs and self-built drones--from passenger carrying aircraft. Meaning rather than sending commands causing drone traffic "swarms" to self-separate or to prevent them from flying outside of their desired area--which strikes me as problematic if you want ot use a drone to inspect a pipeline or inspect telephone wires (for example)--wouldn't it be better to simply create advisories to help separate drones from passenger airplanes?
For example, wouldn't it be better to simply advise drone operators to keep their aircraft 400' AGL (or lower) and keep them out of the approach corridors of various airports (and publish those locations and encourage drone manufacturers to provide maps), and pass laws which make it a potentially criminal offense to operate a drone above 400' or in a landing corridor unless your drone has a transponder, you're in constant contact with ATC (and are being actively separated by ATC) and the drone has the ability to land itself when communications with its operator is lost?
Yes, it'd be a bitch if two drones collide. But there you're just talking about property damage. What frightens me is some idiot with a DJI Phantom and a GoPro trying to get a close-up of a 737 carrying passengers as it attempts to land at LAX--and getting his toy sucked into the engine, taking that engine out and risking everyone on board and everyone on the ground.
This is just more government and not something we need. People who think the skies would be packed with drones have no concept of 3D space. Drones are tiny. Skies are huge. Value of drones is low. Air Traffic Control for drones is totally unnecessary. This is just a distraction from the much more important issue of privacy.
implement flock vs flock avoidment behaviour.
This will likely cost a lot to use: a competitive market for 'transactions' and licensing. Imagine each segment or corridor of airway being owned and sublet by someone who sets transit pricing. Imagine the licensing process itself being regulated like domain names. It's likely to be better if regulated exclusively by a central authority, on a not-for-profit basis.
It's just a shared 4D data structure. With something like a few thousand users and perhaps up to a few dozen transactions per second at most. Why should it be impossible to do in a central location? It should even scale nicely into geographic cells for distributed transactions if you actually needed that.
Ezekiel 23:20
Before they can do that, they'll have to open a chain of pediatric and veterinary hospitals.
I like the unwritten but implied bit about all drones having code which allows them to be controlled by this central system, at a minimum to be forced to land.
Regulations will come out stating that all drones have to have Airware software running on them allowing the central control system to be able to land them or modify their flight plans in case of a need. Any drone found flying without it will be free game to bring down via other methods and/or subject to a fine and loss of the drone to the government.
Eventually, it will be a thoroughly regulated and controlled system, with an every day Joe Shmoe unable to afford the drone anyway with all the registration fees and insurance requirements.
No one needs to control whole flock. Individual birds only need to follow what nearest neighbors are doing.
Why does it need to be centralized at all? The only way a collision is going to occur is if two drones are trying to fly through the same piece of sky at the same time. If they're close enough to do that, they are more than close enough to talk to each other wirelessly. They can then autonegotiate between themselves which one yeilds to the other.
When our name is on the back of your car, we're behind you all the way!
So you're basically asking for expensive specialized hardware for just a few thousand users? That solves something that might not even be an issue?
Ezekiel 23:20
No. I'm recommending a cheap localized solution vs the expensive bureaucratic centralized solution that you're suggesting. Why should I have to talk to Washington to fly a drone over my own field?
When our name is on the back of your car, we're behind you all the way!
Apparently I should've leaned less on snark in my original comment.
By mandating central control, you're making so many assumptions -- the central controller is correct, the central controller scales successfully to the maximum traffic level, there is reliable communication at all times between the central controller and every autonomous agent, every autonomous agent correctly reports its position and status to the central controller, every autonomous agent responds correctly to direction from the central controller, and those are just off the top of my head.
I think the odds of getting every one of those elements right are vanishingly small, compared to "each autonomous agent implements collision and congestion avoidance to the best of its ability". This isn't my field, so I may be far, far off base, but I'm honestly not trying to troll here...
A perfect use for Cloud Computing.
They can take my LifeAlert pendant when they pry it from my cold dead fingers.
I think they are way off the mark as to the nature of the problem they are trying to solve as well as the timing of it. I believe there is a good possibility that we will have not thousands or tens of thousands, but millions of smaller drones in the air by 2020. And, yes, they will be autonomous or semi-autonomous. Putting Google cars on the road is harder than making drones autonomous.
The key to understanding here is that these are just robots. As we move further into the age of personal robotics, there will be many many tasks that a robot that can fly will be better able to do for us than a robot that is limited to walking or rolling around. Many of these devices are also very small. Once they become quieter and smart enough to auto-fly through hallways and crowds, I see no reason why these devices wouldn't go everywhere we go. I foresee them flying through doors into buildings, possibly switching into and out of rolling modes and delivering items right to a person, not just to a building. Or, to get away from the delivery theme, they could be flying around picking up trash, washing windows, getting leaves off of roofs, trimming trees, stringing poles so that nobody has to climb up them, changing lights on towers, flying the rounds of a security guard (even through halls),,, who knows, disposable cameras might even have the ability to sprout a prop, fly off 15 feet, take your picture, come back and land on your hand (there was a bracelet that did something like this on YouTube recently). It's all moving fast enough now that none of this is unrealistic.
So, given that it would take the government 15-20 years to deploy a system, it will be way too late. The need, the explosion of devices, will come from the home-based, personal uses, not commercial businesses, and it will come in the next few years. I've already seen devices that automatically launch, perform a chore, dock, and recharge. It's one of the next big things, and it is way closer than people think. If companies don't do it for us, we're going to do it ourselves.
These systems are going to have to control their traffic the same way people do, with eyes, ears, and some rules of the road. They are going to be interacting and intermingling directly with us, not just each other. Even thinking about centralized control is just a way to subsidize some scientist who would do better economically spending his time designing these devices himself instead of telling others how to control them. We're not waiting for his advice.
The parent's post is wrong on so many levels. Let's start with the technical:
1. Most multi-rotor toys which are bought from anywhere more dedicated to the hobby than toys-r-us ARE drones. Can it go by itself? A lot of people will answer no, and yet when their controllers drop out the drone (and I will keep using that word) will return to launch, or hold position. Most entry level drones are controlled via GPS and barometrics. That is how the hobby has suddenly taken off to thousands upon thousands of unskilled fliers. The ability for it to not just fall out of the sky when you're not paying complete attention to me implies a certain amount of autonomy.
Then there's the hobby and the use of the devices themselves. Yes I fly mine with controls, sometimes I even fly it fully manual. But typically when I finish I just flip a switch and meet my drone back at my car. My drone was hyper expensive, a whole whopping $500. Yep that's right, it's usually one of the cheapest toys in the local park. But that is how I CHOOSE to fly it. I can just as easily load up software on my phone and via telemetry send it a flight path and hit go, but where's the fun in that. Done it once and it was boring. It doesn't make my drone any less of a drone when I send it commands continuously.
2. Then there's your grasp of English. Despite what you think drone means, we have dictionaries for a reason:
drone noun (AIRCRAFT)
a type of aircraft that does not have a pilot but is controlled by someone on the ground
But even if we ignore the dictionaries, most of these toys over about $120 come with some sort of FPV system now. The FAA defines these as drones too.
Just because the word drone started off as expensive military toys, doesn't mean that a device which has the same features (remote operation, most drones were never autonomous) isn't worthy of the title.
Why Washington? The whole thing can be geographically sharded, in fact. Also I don't see why one should call fully automated systems "bureaucratic". Is it any more bureaucratic than a P2P solution with custom hardware, or a lot of things running in your computer? That would have to do a lot of "bookkeeping" anyway. Are transactional database systems "bureaucratic" because they resolve access conflicts? I wonder why we're still using them, then...
Ezekiel 23:20
That's right, these people are just pretending they're in a real helicopter. IS that so wrong? I mean the stuff of little boy dreams is now possible.
It is if you get in the way of an actual manned helicopter. If drone/RPV usage keeps going up regulation will be needed, sooner or later. A good part of a private pilot license exam is regarding regulations and air traffic rules. Even if you only plan to spin around a runway in a small Cessna these still apply.
Imagine what would happen if people started putting radio controlled cars in a highway. Now imaginge the highway is 3,000FT above the ground.
Why Washington? Because The Fancy Article is talking about filing flight plans, and since a central authority has to have a center, and Washington is the location of the central controlling authority in the US. (I suppose I should have said Warrenton, Virginia, home of the Air Traffic Control System Command Center, but I didn't know that earlier).
Why do you want a central controlling authority in the first place, whether or not it is geographically sharded? Why is there any need for any sort of database, 4D or otherwise? Simply have the drones talk to each other and negotiate how to avoid each other directly, like small plane pilots following visual flight rules. No need for a database. No need for any authority.
When our name is on the back of your car, we're behind you all the way!
Because The Fancy Article is talking about filing flight plans, and since a central authority has to have a center, and Washington is the location of the central controlling authority in the US.
It doesn't have to be a federal center; in fact, it makes little sense to have one for that.
Why do you want a central controlling authority in the first place, whether or not it is geographically sharded?
Because hardware-wise, it's simpler to implement (basically COTS), with potentially lower amortized costs.
Why is there any need for any sort of database, 4D or otherwise?
Because computational geometry is the fundamental nature of the problem, from which the structure of the solution naturally arises. Just like using wheels is a natural solution to moving things on the ground (to make a car analogy). But of course you'd suggest mechanical limbs, because that's what apparently seems more natural to you.
Ezekiel 23:20
It seems more natural for one drone to tell the other "get out of my way" than for that drone to phone in and say "Hey Central, contact drone in sector XYZ and tell it to get out of my way".
When our name is on the back of your car, we're behind you all the way!
We (the responsible R/C helicopter community) have been asking, in fact begging, for regulation for some time now. Even going so far as to have pseudo-regulation via AMA. Unfortunately the FAA has continually put the issue on the back-burner....
I'm aware of this. Sadly, this won't change until there's money involved, like Amazon's recent research on using quads for package delivery. These regulations are strict and hard to change, but, again, there's good reason for it.
Also, the analogy of radio-controlled cars is absurdly non-sequitur as cars are 3,000lbs of steel/aluminum and multi-rotors are 27 oz. of plastic/carbon fiber.
The thing is, you don't need a lot of mass to damage something that flies. A RC car can wreck havoc if stepped over by a car going 70MPH on a highway, just like a quad can do even more damage if sucked into the jet intake of an airliner / helicopter. Or the propeller / attack surfaces of a small airplane - incidents with birds, for example, are sadly common.