DOT and FAA Propose New Rules for Small Unmanned Aircraft Systems | Hacker News Reader

DOT and FAA Propose New Rules for Small Unmanned Aircraft Systems

faa.gov

66 pointsghaff30 comments

The proposed rules seem prudent and well-considered. I'm an amateur radio operator, and there are similar licensure, testing, and regulatory requirements in order to ensure the best use of a shared and finite resource. Where radios have restrictions on frequency, aircraft have restrictions on location, where transmitters are prohibited from causing interference, aircraft are controled for safety. Where radio power is limited (to the the minimum necessary for communication), aircraft use is limited to line-of-sight. Most of the populace is unlikely to be an operator (of radio, or aircraft) so it is important that those who are are respectful of everyone else such that their use does not endanger or interfere with ordinary activity.

Just as no one wants an unlicensed transmitter blocking the ability for a cell phone to call 911, no one wants a quad copter falling on his or her head.

I thought a while ago that ham radio regulations might be a useful conceptual template and, as you say, that's what seems to be happening.

I imagine we'll see some pushback on the requirements for getting certified (e.g. background check) but that's quibbling at the margins.

I also imagine that some won't like the prohibition on autonomous operation outside line of sight. But that seems a reasonable restriction to me if we're heading toward a spot where there are potentially lots of drones flying around.

I heard some of the information that came out of the rule making committee about this time last year. The FAA has the idea that they need to control this as fast as possible, while doing what they can to slow the explosion so that it remains under constant control.

Part of me wonders if the background check (TSA Screen) is just something to weed out people who are on the fence about getting the certificate.

I do applaud some of what they're doing except for the weight classes they have assigned. The 55 lb and under is a huge range. They are making safety rules based on a contraption between the weight of a desktop computer or a toaster. I'm not sure a toaster falling on your head from 500 feet would kill you, but a computer falling from that height surely would.

Well, a tape measure that dropped 500feet killed a man[1], so I wouldn't be too confident about something the size of a toaster not killing someone when being dropped 500ft.

Granted a tape measure does weigh less than 2lb (~toaster), but anything dropping from 500feet is still cause for concern.

[1]http://www.krmg.com/news/news/national/tape-measure-dropped-...

I wasn't confident and said I wasn't sure.

Maybe the rules according to those weights aren't wrong.

I wonder what the terminal velocity is for a drone that's lost power. It seems like anything that's able to climb to and hover at 500 feet probably won't drop quite like a brick unless things go wrong in a very particular way.

This looks reasonable. A 25Kg drone is a serious threat to aircraft, and, reasonably enough, the FAA wants to make drone operators take a test and get a license. UAVs in these classes have to be operated line of sight from the ground operator and under 500 ft AGL, and only in class G (uncontrolled, usually below 1200 feet) airspace.

There's a comment that "operations in Class B, C, D and E airspace are allowed with the required ATC permission." What that basically means is no drones anywhere near an airport, where class B, C, and D airspace goes all the way down to the ground and a drone under 500 feet might be in the way.

"Near an airport" covers a lot of territory near major airports. See "http://skyvector.com/". Wide blue lines show airspace boundaries. Where you see 100/SFC, that means the controlled area is from 10,000 feet to the surface. Big chunks of metropolitan areas are thus no-drone areas.

If you really need to operate a drone in controlled airspace, it's going to require all the usual procedures for flying in controlled airspace - a flight plan, radio contact with ATC, and probably a transponder so the controllers can see the drone on radar. Sagetech has a 100 gram transponder for UAVs which identifies a drone to ATC radar, just like larger aircraft. So when your drone is taking pictures of a building, ATC can see that it's at 250 feet, sending the ID they issued you with the flight plan, and nowhere near a runway approach. They're in contact with you if they need you to move the drone. This is a lot like helicopter operations in urban areas. This will cover the needs of movie companies, bridge inspectors, and such.

There's a proposal for comment mentioned for a 2Kg "microUAV" class with less regulation. That's useful, and will encourage toy manufacturers to stay below that weight limit.

None of this allows flying drones beyond visual range. The FAA isn't taking on that problem for now.

I'm sure there will be a lot of back and forth on the rule specifics, but seems like a reasonable first pass given the state of the technology.

The only thing that really strikes me as a bit problematic is "casual" commercial use in which someone like a real estate agent uses a hobbyist-type drone to do a little aerial photography. These rules seem a bit stringent for something like that although the reality may be the rules for commercial use just get ignored at these kind of margins.

- An FAA UAS operator certificate will be required for commercial use. I don't think the details of what this will entail have been worked out yet but it explicitly will not have the same level of rigor as a pilot's license.

- Requires the operator to maintain line of sight to the drone.

- Can't fly over people not involved in the flight.

- Certification isn't required for hobbyist use.

- Considering a more flexible framework for drones under 4.4 pounds.

What about property owner permission? I feel like that's something really important. Even if I'm not at my house I should still have a say if someone wants to fly some 50 lb contraption over it.

Well you don't own the airspace over your property. Based on existing regulations that's class G and is open and uncontrolled (generally from the ground to 1200' AGL)[0]

And if any of the regs in part 91 end up applying-

FAR 91.13 (a) just says aircraft have to be operated safely. "No person may operate an aircraft in a careless or reckless manner so as to endanger the life or property of another." So unless they were buzzing your house that might be hard to prove.

The regs are loose around dropping items, too- FAR 91.15 [1]. They don't talk about needing property owner permission or anything. Maybe you could twist that in your defense since it is so loose- if something did fall from a UAV.

[0] http://en.wikipedia.org/wiki/Airspace_class_%28United_States...

[1] http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgFAR.nsf...

FAR 91.119 [1] Proscribes what you describe. Flight below 1000' from the highest obstacle within 2000' horizontal is disallowed in congested areas (cities, towns, etc), and disallowed below 500' AGL in other than congested areas. Clearly small UASs are going to be exempt from those parts akin to the exemption for helicopters in subsection d. But there is no exemption for any aircraft from subsection a, so I'm curious how commercial or hobbyists comply with that.

[1] http://www.flightsimaviation.com/data/FARS/part_91-119.html

I think FAR 91.119(a) which I mention below applies here. The thing is, aircraft have registration requirements with the FAA so tracking down the owner/operator is straightforward. There needs to be something similar for UASs, and owner/operators will want to be really clear establishing who the operator was at a particular moment, and who has liability. That's something companies and their insurance companies will sort out over a relatively short amount of time. I'm more concerned about hobbyists than businesses at this point. Businesses have more to lose, not just lawsuits, but seriously bad imagery if someone's hurt by one of these things. (55lb doing 100mph from 500'? OK then! Look out below!)

The point is probably that there should be no casual commercial operators. If you need some aerial photography done, hire a qualified drone pilot for a few hours. In fact, that will probably become a specialty business of independent operators or small companies - qualified pilots with high-quality drones and camera gear. I wouldn't be surprised to see this become a "pivot" for current commercial photographers in the weddings/product photos business.

Depending on the complexity of your project, that might cost more than outright purchase of a cheaper(er) drone, but I would argue this is a net win. A whole little business niche gets created, and most of the drones in the sky are 1) flown by experienced operators, and 2) high-quality and therefore probably easier to control.

> "This looks reasonable."

I'm a pilot, so familiar with FAA rules.

The first 3 bulletin points are reasonable, but the last 3 are onerous for UAS operators. Not flying over people will mean not flying in cities, the point of commercial operators like Amazon.

Also, the FAA likes to grant licences so that then they will have power to revoke them at will. Fighting an FAA certificate revocation action is like trying to get off a no-fly list.

In a recent podcast by Adam Savage[1], an avid drone enthusiast, drones are described as "cinder blocks with propellers attached" or something to that effect. In one incident described in the podcast, Will describes a multiple-redundant multirotor losing a single propeller and having the vibrations of that cause a cascading failure of the other propellers, ultimately bringing the machine down in an uncontrolled manner.

I think that as a blanket rule it is very appropriate to not allow amateur UAS to fly directly overhead of people not aware that it is there. If a machine should be built that is provably safe against such failures or demonstrates an extreme tolerance to them, then special licensing could be dispensed.

[1] https://www.youtube.com/watch?v=0RvHI79CyqQ

> Fighting an FAA certificate revocation action is like trying to get off a no-fly list.

That just isn't true.

People have actually gotten off the no-fly list!

https://www.aclu.org/blog/national-security-technology-and-l...

The last three are oppressively burdensome? You think a UAS should be able to fly above 500' or above 100mph? And fly in airport flight paths and restricted airspace? Why do you think it's reasonable for a piloted aircraft to be disallowed from flying below 1000' from the highest point of any obstacle within 2000' from an airplane, but you think it's OK to fly an unmanned done well below this altitude above 100mph right over people? That makes no sense. Please clarify.

Note especially that the proposal is for "non-recreational operations" - FAA tends to make a big differentiation between commercial and private (recreational) activities. I read this proposal as being aimed at movie production, utility (pipeline, powerline) inspection, and similar operations.

Another thing about the FAA - once they write rules, they are generally reluctant to change them, so for anyone wanting to comment on the NPRM, sooner is better than later.

These rules are very similar to those already in place in Australia (http://www.casa.gov.au/scripts/nc.dll?WCMS:STANDARD::pc=PC_1...).

55lb going 100mph rapidly descending from 500' AGL. Hmm.

"The new rules would not apply to model aircraft." I think this is going to be a problem... when does a model become a "small unmanned aircraft system" and when does it not? It seems to me they're at least heavily overlapping, if not synonymous.

Does it have to resemble an existing aircraft to be considered a "model aircraft"? This could easily be open to interpretation of law enforcement. If so, my brother had some "small unmanned aircraft systems" back in the '70s (powered by 049 engines). They were most definitely "experimental designs" as some of them looked nothing like any (manned) aircraft ever produced. This proposal looks dangerously close to something that could make experimenting with new aircraft designs an illegal activity, including those designed for humor. Examples include flying lawnmowers, doghouses and witches. [0] How would they classify my Dragonflies? [1] They certainly don't resemble any manned flying contraption ever made, they barely resemble an actual dragonfly. Would these fall under the "more flexible framework for “micro” UAS under 4.4 pounds"? If so, that's utterly absurd. If your answer is "it's a toy", I agree, but then so are all of the quads and aircraft I've already mentioned. It needs to have a very distinct line distinguishing a "model aircraft" from a "small unmanned aircraft".

It doesn't even say "powered aircraft", so a large hand-launched remote controlled glider could be illegal (yes, I know this is reaching a bit, just trying to prove a point).

It's unclear to me if this will make illegal all of the automated fliers that have been built (such as those suggested on old episodes of Hak5, programmed to fly to a nearby building's roof, park for a period of time, then fly back, etc.). I suspect it will, since it makes repeated references to the "operator" and "line of sight", etc. I saw interviews with people (at DEFCON?) who had created some very impressive auto-navigating (beyond autopilots) craft as well. All of the machines produced with cameras and video feeds for remote operation would suddenly be illegal to fly in the manner for which they were designed. Among those would be a wonderfully implemented one with a full cockpit view. [2]

"A small UAS may not fly over people" This is really vague... "directly over" or just "kinda over" (what range)? Technically, if they're over 7 feet up, they're "over people" so already in violation. This will make flying remote control airplanes illegal in parks too... where will all the hobbyists be able to go? You can't launch and land in a cornfield.

[0] http://flyingthingz.com/ [1] http://www.wowwee.com/en/products/toys/flight/flytech/dragon... [2] http://hackaday.com/2014/06/18/rc-plane-flies-with-a-cockpit...

To me this is similar to old red flag laws requiring a motor vehicle to be preceded by a person waving a red flag.

For fully autonomous vehicles, these US regulations appear to be a barrier. How could merchants pursue autonomous drone delivery?

Perhaps some other jurisdictions, where entrepreneurs have more freedom to innovate with this tech - will show the way.

Seriously? There's nothing to keep companies from working to figure out what the challenges/requirements/etc. are for doing home delivery at some point. I also suspect it is quite a long way away and, in the meantime, I don't want to see random companies going all move fast and break things with heavy objects flying through the air above people and in controlled airspaces.

Venture capitalists and self funded R&D want rosy market projections.

Sadly, based on sentiment from USA residents, like yours I suppose, I predict the USA is going to lead with military applications only for UAVs.

> Venture capitalists and self funded R&D want rosy market projections.

I want people to not die because somebody "moved fast and broke something."

I wonder how many lives would be lost due to failing drones vs lives saved by the timely delivery of medications/defribillators by waypoint-following autonomous air vehicles.

The data will most certainly come either from US military usage or from commercial usage outside the USA. Sad.

There's nothing stopping your hypothetical company from getting FAA permission for a trial, even in an urban area. Yes, it requires one of those "lawyers", and requires you to have more of a plan than just "eh fuck it, light it up," but those are features, not bugs. But nobody's going to be prevented from doing even urban-scale testing if they are actually taking it seriously and involving the communities they wish to experiment within.

There's a reason these rules look so much like radio regulations: because screwing it up hurts everyone, so it behooves us not to screw up. Doing dangerous things requires buy-in from more than just your shareholders. And stop with the "oh, woe is us" stuff. It's shitty rhetoric and it doesn't work.

Well we are talking past each other.

I predict that timely delivery of life saving medications and medical devices will first happen outside of the USA, in jurisdictions favoring UAV innovation.

http://www.uasvision.com/2015/02/10/japan-tests-drone-delive...

http://en.people.cn/n/2015/0209/c98649-8847897.html

And I predict that the in the USA, the military will lead the way with regard to autonomous air vehicle technology for delivery because not the FAA, nor local equivalents, will constrain them where battles are happening.

http://www.defensenews.com/story/defense-news/blog/intercept...

Autonomous drone delivery requires something analogous to "positive control" used in the existing air traffic control system, yet none of that infrastructure exists. Neither primary nor secondary radar used for aircraft today will work on UASs, so how are they going to be tracked and separation ensured? If it's a decentralized system along the lines of TCAS, that implies some industry agreed upon spec, whether it's active/passive sensing, or whether there's actually communication between UAS's. Any communication probably needs to be cryptographically signed for validation, probably shouldn't be encrypted however so it's transparent. Clearly it has to go beyond existing technologies for aircraft if there will be no role for air traffic control. A whole separate set of expectations and standards for how various kinds of systems failures are to be treated, how does an autonomous UAS determine an emergency, how does it declare one (to reserve a different section of airspace than on its prior "plan" in order to avoid a worse accident), how does it actually perform an emergency landing? Some of these things could be proprietary but other things need to be openly published, namely the spec, the standard, and the metric for performance.

We can't just say "oh, no pilot, rules go out the window" as if everything we've learned about flight safety doesn't apply at all to UASs. No, there's actually a lot of adaptation that needs to be done. A whole lot.

Yep. The US military and their civilian/academic researchers are working on this as part of the future battlespace. The 2015 US military budget has $700 million for drone procurement and R&D.

Reply on news.ycombinator.com