It was a ‘technical problem’ that sent my DJI Phantom towards a pine tree. Ok, so maybe I was chasing my nephew with it, and things got out of hand. My nephew had a blast while I had a slight heart attack as my Phantom hit the ground. Luckily, it came out unscathed.

New research is out from RMIT University’s School of Engineering and took a more in-depth look at the cause of drone crashes. It turns out, crashes are more likely the result of technical problems than human error. Does me not recharging my battery count as a technical problem?

DJI box open

Researchers looked at more than 150 reported civil incidents around the world between 2006 and 2016. In 64% of the crashes, they found technical problems were to blame – not the drone pilot. The issue with assigning blame to technical problems is we don’t know the circumstances of the incident. Was the pilot flying outside the technical/legal parameters for the drone?

The biggest issue, according to researchers, centered around broken communications links between the pilot and the drone. Dr. Graham Wild says drones with more “robust communications systems” could help prevent accidents. The research team wants to see commercial aircraft-type regulations govern the communications systems.

The aircraft we fly when going on vacations have triple redundant systems for communications.

“But drones don’t and some of the improvements that have reduced the risks in those aircraft could also be used to improve the safety of drones,” Wild added.

One problem according to Wild is how much regulations differ based on the size of the drone.

“Drones are being used for a wide range of tasks now and there are a lot of day-to-day activities that people want to use them for – delivering pizzas and packages, taking photos, geosurveying, firefighting, and search and rescue,” says Wild.

“It’s essential that our safety regulations keep up with this rapidly-growing industry.”

Redundant communications system is a smart, and simple, way to make drones safer for everyone. There are things all of us can do to make drones safer but still have fun. Manufacturers can tighten up the technical problems and we (pilots) can fly smarter. Or, just use common sense. Don’t take drones near airports and stay out of National Parks.

At least now you have an excuse for when your drone clips a tree limb. Or, you manage to flip the drone upside down when landing it. I don’t know how my dad did that last one, but he found a way.

Update: DJI is pushing back hard against the study, questioning the methodology and conclusions in the report. And they have a point. Me flying a Phantom into a tree isn’t a technical glitch. It’s me chasing my nephew for the hell of it. 

The recent study on drone incidents published in Aerospace is laughable. It lists just one example of a drone incident, involving a jet at Heathrow – but British aviation authorities say it was most likely a plastic bag, not a drone (http://www.independent.co.uk/news/uk/home-news/drone-british-airways-plane-plastic-bag-london-heathrow-a6996031.html).

The researchers claim to have searched 19 data sources to find 152 cases of drone accidents or incidents from 2006 to 2015, yet the FAA alone receives more than 100 reports of drone sightings every month. (http://www.faa.gov/uas/resources/uas_sightings_report/) The most rigorous analysis to date of those reports found that only a tiny fraction indicated a “near miss” or “close call,” and many of them appear to have been operated safely. (http://amablog.modelaircraft.org/amagov/files/2016/06/AMA-Analysis-FINAL-6-1-16.pdf) The authors acknowledge finding their examples through “convenience sampling” with “no random sub-sampling.” That seems to guarantee a skewed data set, but since that data is not included in the paper, there is no way to know how they chose “accidents and incidents” to support their conclusions.

There has never been a confirmed collision between a civilian drone and a manned aircraft. No one has ever been killed by a civilian drone. While the authors call for more government regulations to ensure drone safety, regulators themselves have taken the opposite position. Since small civilian drones have a much smaller risk profile than manned aircraft, they have encouraged innovators to develop promising technological improvements rather than subject every new drone model to a bureaucratic approval process. That’s true in Australia, it’s true in the US, and it’s true in multiple other jurisdictions where regulators agree there is simply no need to put small drones through the same rigorous years-long airworthiness process as a jumbo jet. And as for a reporting regimen … do they really think every time someone crashes a small hobby drone into a tree it should get reported to the government? Even if they’re seriously making that suggestion, no regulator would want that flood of reports.

This paper is utterly specious, and its conclusions are implausible. Equipment failures do occur from time to time, and we are constantly innovating technological improvements to prevent them. But problems with drones are far more likely to occur because of operator error, which is why we support educational efforts to ensure all drone pilots learn to fly their craft safely and responsibly.

Translation – The study is a bit hyperbolic in its conclusions and recommendations. No one has been killed by a civilian drone and accidents are more likely to be attributed to operator error – be it charging the batteries before flying and keeping visual contact with the drone.

When I’m not playing Rocket League (best game ever), you can find me writing about all things games, space and more. You can reach me at alex@newsledge.com

You may also like


Comments are closed.