Originally published at: Drone Collisions: A Growing Risk - AVweb
Frankly, I haven’t worried much about running into a drone as much as other aircraft. I had a close one in an uncontrolled traffic pattern in Florida recently where a guy in a Cherokee dropped onto the downwind—from above—and then broke off and never said a word to three of us established in the pattern.…
Drone use is exploding, recreational, commercial, legal, illegal, smart and stupid. The skies are getting crowded, and the threat to aviation is no longer theoretical. Since 2014, drones have been linked to over half of all near-midair collisions at the 30 busiest U.S. airports. In 2024, nearly two-thirds of near misses during takeoff and landing involved drones. This year alone, over 3,000 drone incidents have been logged near airports. Aircraft have taken evasive action. Firefighting and medevac flights have been disrupted. One misjudged drone flight could kill dozens.
The FAA’s Remote ID rule was supposed to help, but it’s been delayed and ignored. Many drone pilots fly beyond line of sight, disable safety features, or simply don’t know the rules. Meanwhile, law enforcement is under-equipped, and public education is weak. We’re gambling with lives at 3,000 feet.
This isn’t about banning drones, it’s about enforcing accountability before tragedy forces our hand. Mandate Remote ID. Penalize violations. Require training and registration. Launch national campaigns to teach safe flying. Drone freedom without responsibility is a threat, not a right.
Pilots operate under strict regulation. Drone users must too. One crash will rewrite the future, for everyone. Let’s act before that happens, not after.
If the FAA doesn’t get ahead of this, and fast, we’ll end up with a fractured, corporatized, and unsafe low-level airspace where drones dominate by default and manned aviation is boxed out by risk, cost, or regulation.
You’re either managing the airspace, or you’re losing it.
Sorry, but what you’ve posted isn’t accurate at all. While the FAA has kept a database and does quarterly reporting on what is termed “near miss”, the list has no validation or verification but is instead just a compilation of reports. It has long been criticized for being the new “I see a UFO” report. Many of the reports are very clearly NOT drones at all. Here is a great example quoted right from the FAA data:
" Pilot reported a drone or large mylar balloon overhead the airport at approx. 3500 MSL. The pilot described it as round, a balloon in the center with things hanging off of it, unsure but possibly numbers like from a birthday balloon. There was a main body with protrusions, light or metallic grey in color. The size was approx. 3 ft across and 2 ft in height. The pax reported watching a departure from SAN and then this attracted his attention. No evasive action required. They reported going into slow flight to try and see it again with no success."
In a 2015 review of the data the AMA found only 27 line items where the pilot suggested a “near miss” of the 764 that were in the report.
It is also worth noting that the ALPA actively tells its members to report drone sightings, yet it simultaneously claims drones are “extremely difficult” to see. A study done by a Colorado Ag group found that low flying helicopters failed to see drones flying in the same airspace even when provided visual cues (colored flags in the grid alerting of a drone within the same grid).
Now to be quite blunt: Drones are the safest aircraft in the skies, bar none. The last fatality anywhere in the world from an sUAS was in September of 2013 - Roman Pirozek Jr. was killed when piloting his RC helicopter and having the rotor hit him. Not a single fatality since. The number of significant injuries due to sUAS over the past dozen years is also minimal. Compare this to manned aviation where in the US alone helicopters account for an average of near 20 fatalities annually, airplanes even more, and the property damage from those manned aircraft total in the tens of millions annually. You don’t find that with drones.
For manned aviators concerned with mid-air collisions, understand that such will happen and has. But the risks are exceedingly small and have been extensively studied. I highly recommend a read of the ASSURE reports that were commissioned by the FAA on such.
1 Like
@UAS_Legal: You are framing this like a legal debate, but the operational reality is clear. Drones pose real collision risks to helicopters and other low-level aircraft, especially during takeoff, landing, and emergency operations. The FAA and its ASSURE program have acknowledged this risk.
Calling the FAA’s near-miss database a “UFO log” ignores verified incidents. The 2025 Palisades Fire drone strike tore a hole in a CL-415’s wing. That incident was confirmed, investigated, and prosecuted. The LAPD helicopter collision was real. So were multiple disruptions to firefighting operations caused by drone interference.
You cite the 2015 Colorado study to discredit pilot reports, but it shows the opposite. Pilots failed to see drones even when warned and when visual aids were used. That is not a flaw in reporting, it is proof that drones are often undetectable in real-world conditions.
Claiming drones are the safest aircraft because they haven’t caused a fatal midair misses the point. Safety is about preventing loss, not waiting for it. Manned pilots have avoided drones so far, but that margin is shrinking.
ALPA encouraging drone sighting reports while acknowledging visibility limits is not a contradiction. It reflects the reality of incomplete data and responsible safety reporting.
Your argument rests on limited visibility, inconsistent reports, and a lack of fatalities. These are not signs of safety. They are early warning signs, just like the ones aviation has ignored in the past. The absence of tragedy is not proof of safety. It is proof we are burning through luck.
Safety isn’t the elimination of risk. If it were, then no aircraft would be allowed to fly, no person would be allowed to ride in a car, no movement would be allowed. The signs of safety are years worth of data. In the case of sUAS that is more than a decade. Name any decade in the more than 12 decades of manned flight in which there have been no fatalities. Mitigation of risks and reasonable regulation are what bring about safety. For manned aviation the rules and regulations have been “written in blood”, a retroactive response to accidents. For sUAS the rules and regulations were proactive - thinking ahead of time about what mitigations were necessary to allow sUAS and keep risks low. Over the past decade the FAA has further recognized that the regulations they put in place were overzealous. They have eased regulations as a result, including night flight, flight over people, and will soon ease VLOS restrictions. They do so recognizing that the risks of those operations are mitigated and exceedingly low.
You call out two mid-air collision incidents between sUAS and manned aircraft which aren’t even half of the incidents in the US alone over the course of the past dozen years. Most don’t even know of them as they aren’t significant news. They have been real world examples that have affirmed the ASSURE report conclusions that the risk of catastrophic incident is tiny. Try the same for any mid-air collision between two manned aircraft.
There is good reason that Congress gave nod to sUAS in the 2012, 2018, and 2024 FMRA. sUAS have been operating safely in the NAS for better than 80 years. The AMA was in place with model aircraft years before the creation of the CAA (precursor to the FAA). The FAA didn’t exist until a mid-air collision over the Grand Canyon in 1956 made clear that more needed to be done.
Aviation safety is most often measured by the number of fatalities against the number of flight hours. The goal may be zero, but the reality is that such does take place. And for manned aviation those numbers have been quite good having falling over the course of the last five decades from 6.5 fatalities per million flight hours to just 0.5 fatalities per million flight hours in the past few years. The number for sUAS is zero and the number of flight hours for sUAS exceed those of manned aviation. Those numbers aren’t on inconsistent reports and the scope of their visibility is worldwide. Where you see burning through luck, the FAA takes note of the very real data and has been easing restrictions with the understanding that the amount of risk doesn’t warrant the stricter regulation.
1 Like
@UAS_Legal: You’re framing this like a historical audit, but the threat isn’t behind us, it’s scaling up underneath us. The volume of drone operations is rising exponentially, especially in the low-altitude corridors used by helicopters, air ambulances, firefighting aircraft, law enforcement, and now, delivery drones. These aren’t weekend hobby flights. They are automated, frequent, and increasingly autonomous operations in the same airspace where manned pilots fly at their most vulnerable: takeoff, landing, and emergency response.
Your emphasis on drone safety based on fatalities overlooks the central concern. Safety in aviation has never been defined by body counts alone, it’s based on risk mitigation before loss occurs. That’s how aviation improved after decades of painful lessons. The fact that drones haven’t caused a fatal midair yet says more about the vigilance of manned pilots than the built-in safety of drones.
Yes, the FAA has eased some restrictions, like allowing night flights and loosening VLOS rules. But relaxing rules before real-time deconfliction tools such as Remote ID and UTM (Unmanned Traffic Management) are widely functional and available to manned aircraft is premature. LAANC access and Remote ID implementation aren’t enough when emergency helicopters operating at 300 feet AGL can’t see, hear, or locate an incoming delivery drone on an autonomous route.
You cite the 2015 Colorado drone visibility study to challenge pilot reports but that study confirmed how hard drones are to detect, even when pilots were alerted and the drones marked with strobes and flags. That isn’t a knock against pilot reliability. It proves that line-of-sight detection doesn’t cut it in low-altitude, real-world flying.
The numbers also contradict your sense of proportion. As of 2025, over 865,000 drones are registered with the FAA, more than six times the number of registered manned aircraft. Commercial delivery drone operations are projected to exceed millions of flights annually by the end of the decade. That scale shift alone warrants proactive oversight, not relaxed vigilance.
So who’s really being stubborn? It’s not the pilots flagging operational blind spots. It’s the voices claiming that because we haven’t had a catastrophe, there’s no need to act. That’s not foresight, it’s waiting for the headline. And when you wait for the headline, safety has already failed.
This is not about resisting change or opposing technology. It’s about protecting the airspace where people are already flying, already saving lives, and already sharing risk. Complacency in the face of rapid growth isn’t progress. It’s how safety margins get erased.
Simply put, the FAA is fully dedicated to the safety of the NAS. Your argument is that they are being stubborn. Either the FAA is modifying regulations to come in line with the reality of risk or the FAA isn’t happy until you’re unhappy.
No, the argument is not that the FAA is being “stubborn.” The argument is that the FAA is easing restrictions before the infrastructure exists to safely support that expansion.
Adjusting policy is not the same as managing risk. Remote ID delays, limited UTM visibility, and weak enforcement don’t align with the pace or scale of uncrewed traffic growth.
Pointing that out isn’t about being unhappy, it’s about being realistic. Safety isn’t a feeling. It’s a function of what the system can handle. And right now, it’s behind.
We’re done here.
All you people have done a good job of stating the drone-strike problem and have danced around the solution. However, you have forgotten the second most tightly held secret in aviation: See and Avoid is worthless. It is a fiction, a fraud and, used as justification for VFR, has been part of our FARs forever. The truth is: If you see it in time to avoid it, you probably were not going to hit it; if you are going to hit it, you probably won’t see it in time to avoid it. The reason is rooted in basic animal instinct: danger is detected by motion. A collision course is defined by a constant angle of approach. The deer stands still and stares in the direction of the sound or smell. You raise your rifle and it bolts. Or, you are slowing down for a stop sign at a major street, looking both ways, see no traffic and continue with your “California stop.” As you enter the intersection, there is a horn blast from the right and, if both drivers’ reactions are good, you can make a panic full stop and avoid a collision. The traffic, of course, was slowing at the same rate to make a left turn and their turn signal was blocked by the passenger side blind spot. Lesson: always make a full stop and look twice. (Hard to do. 1-2 mph may be ok.)
The most tightly held secret in aviation back in the 1960s when I started flying was AOPA president Doc Hartranft’s salary.
As a chopper owner/pilot who rarely climbs to FL20, I have a fair amount of skin in this game. Relative to an airplane, I’m pretty maneuverable, but nothing like a UAV.
Generally, the right of way is afforded to the less-agile vehicle, for obvious reasons. Cut in front of a tractor-trailer in order to make your exit and, should you miscalculate and get T-boned, it’s your fault. And it’s gonna hurt you 'way more than the semi. But your stupidity could easily cause the death of the truck driver.
The onus for aerial separation is squarely on the shoulders of the UAV operator. He (or it, if autonomous) can see and hear me from miles away. I do not have the luxury of that kind of warning. Some cities are experimenting with delivery droids that use their streets. It would be interesting to hear what sort of de-conflicting paradigm they are using, and how successful it is. Of course, translating that into 3-D may be a Herculean/Sysiphean task.
Regardless, the present state of the art is not sustainable. Given the explosion of popularity in these “aerial mines”, there will be deaths before the FAA gets around to doing anything about it.
I reject the discussion about drones.
I am obviously part of a “select group of people with a lot of disposable income” and a suckerpunch passion for general and business aviation.
Naturally, I am more likely to click on some stupid advertising; than I am to engage in dialogue with my fellow airman.
Thats all that matters! Click on that damn ad, will ya?
A little late to be in the exploration phase of this issue. I’ve been flying “drones” (hobbyist grade RC multirotors, race quads if you will) for a decade now, they’ve been a widely available hobby item that you can buy on Amazon for about that long now too. I didn’t have any near misses with RC aircraft though, my first near miss was a pattern conflict with someone who wasn’t on frequency while flying the real thing though, 20 years ago during a CAP flight. I hope it’s my last, but it was enough to tell me that seeing and avoiding are wishful thinking.
The risk IMO is with people who do not bother to check regulations or deliberately violate them to get photos or are intoxicated.
Intoxicated people do not think.
Before there were many regulations, some UAVs were snooping into apartments overlooking a park in Victoria BC. Police ‘had words’ with them.
Voluntary compliance of drone operators is effective to some extent but there will always be the clueless or defiant who operate drones unsafely near manned aircraft. The incentive to comply with the rules for those folks should at a minimum be loss of the drone. I speak as a commercial pilot, remote pilot and aircraft owner who has flown forest fires for a state agency and encountered unauthorized drones at the scene of a fire. The most promising defensive technology in my opinion is a ground based system microwave transmitter that interferes with the drone’s command and control link such as the Dronebuster. We can’t rely on the threat of prosecution to deter these people and it’s hard to find the drone operator in most cases even with RemoteID. Shotguns are cheaper but not as safe. The regulations need to change to allow defensive measures more latitude.
This topic was automatically closed after 7 days. New replies are no longer allowed.