DJI Demands Withdrawal Of Misleading Drone Collision Video

Simulation Was Staged Faster Than Both Maximum Possible Speed And FAA Guidelines

October 19, 2018 – DJI, the world’s leader in civilian drones and aerial imaging technology, today
demanded the University of Dayton Research Institute (UDRI) withdraw a misleading video and
blog post which claim to depict a collision between a DJI Phantom 2 drone and the wing of a small
airplane.

UDRI staged its video to create a scenario inconceivable in real life, at a higher speed than the
combined maximum speed of the drone and airplane, which is also faster than U.S. Federal
Aviation Administration (FAA) testing guidelines. UDRI has not disclosed its testing methodology or
the resulting data, and while it acknowledged that a similar test with a simulated bird caused
“more apparent damage,” it has only promoted the video showing damage from a DJI drone.

In a letter to the lead researcher involved in the video, DJI’s Vice President of Policy & Legal Affairs
said UDRI “recklessly created and promoted a video that falsely claims to depict a dangerous
condition posed by one of our products. … Your public comments deliberately present an entirely
improbable, if not impossible, event as a commonplace risk routinely faced by airplane pilots.”

The full text of the letter is reproduced below, and a pdf of the letter as delivered is available
at this link.

October 19, 2018

Via Electronic Mail

Kevin Poormon

University of Dayton Research Institute

300 College Park

Dayton, OH 45469

Dear Mr. Poormon:

I represent DJI, the world's largest manufacturer of small unmanned aircraft systems, commonly
known as drones. We lead the industry in developing systems to help ensure drones continue to
safely share the airspace with traditional air traffic. DJI takes aviation safety seriously. It is integral
to who we are as an organization and as aviation professionals. We have proactively incorporated
dozens of safety features into our products, including altitude limitation, airport geofencing,
return-to-home failsafe systems, computer vision anti-collision sensors, and pilot knowledge
testing. We also support research professionals who work alongside the industry and regulators to
provide academic grounding to aviation safety efforts.

It is thus distressing to see how the University of Dayton Research Institute has recklessly created
and promoted a video that falsely claims to depict a dangerous condition posed by one of our
products. Your “Risk in the Sky?” video, blog post and media tour created a collision scenario
between a drone and an airplane wing that is simply inconceivable in real life:

Your video assumes a Mooney M20 light aircraft is flying at its maximum possible speed of 200
mph, and encounters a drone apparently flying faster than its maximum possible speed of 33.5
mph. The plane could only achieve such speed at full cruise, typically more than a mile above
ground. At the altitudes where that plane would conceivably encounter a Phantom drone, it would
fly less than half as fast — generating less than one-fourth of the collision energy.

Your video was created contrary to established U.S. Federal Aviation Administration (FAA) crash
test parameters, which assume a bird striking an airplane at its sea-level cruising speed — which is
typically 161 mph to 184 mph for Mooney M20. Your video deliberately created a more damaging
scenario, and was widely cited as evidence for what could happen to a large commercial jet —
even though the Mooney M20 is a small plane with four seats.

Your video was not created as part of a legitimate scientific query, with little description of your
testing methodology and no disclosure of data generated during the test. Your blog post describes
a similar test performed with a simulated bird that caused “more apparent damage,” but your
decision not to post or promote that video indicates your bias toward sowing fear. This contrasts
with the reputable research performed by the Alliance for System Safety of UAS through Research
Excellence (ASSURE), the FAA Center of Excellence for Unmanned Aircraft Systems, which
meticulously tests a variety of impact scenarios in order to provide the public, the FAA, and the
UAS industry, with supportable conclusions about risk. You have done nothing of the sort.

Given UDRI's wide-ranging publicity efforts in print, broadcast and online media, it seems clear
that your misleading video and incendiary blog post seem designed to generate paid research
work for UDRI at the expense of the reputation of drone technology broadly, and DJI's products
specifically. Your public comments deliberately present an entirely improbable, if not impossible,
event as a commonplace risk routinely faced by airplane pilots.

To elaborate on the points outlined above, the impact velocity tested, 238 mph, far exceeds any
conceivable collision speed between a Mooney M20 and a DJI Phantom 2. The M20J Pilots
Operating Handbook lists the maximum structural speed of a Mooney 20 at 174 knots, which is
200 mph. Cruise speed will typically be 140-160 knots (161-184 mph), more than a mile above
ground. The Phantom, and our other drones, have built-in altitude limitation features. Thus in the
altitudes no higher than several hundred feet above ground where a drone is likely to operate, the
Mooney M20 would be taking off or landing at speeds between 70-88 knots (81-101 mph).

As for the other aircraft in this scenario, DJI has not manufactured the Phantom 2 drone for years,
but its published specifications indicate a top speed of 15 meters/second, or 33.5 mph. In other
words, it is virtually impossible for these two aircraft to encounter each other at the speed of your
test. Given that kinetic energy, and therefore resulting damage, increases by the square of
velocity, the arbitrary increase in your test velocity results in dramatically more damage.

More to the point, a test deliberately designed to generate the worst conceivable outcome is
contrary to the FAA's established testing parameters, which seek to measure the risk that an
aircraft is most likely to encounter. The relevant Federal Aviation Regulation states an airplane
must be capable of successfully completing a flight despite striking a bird at the equivalent of the
aircraft's cruise speed measured at sea level, which as stated above is 161-184 mph for the
Mooney M20. Your test was thus performed at a speed 54 mph to 77 mph faster than a
responsible collision test would require, creating a case that is unrealistic and damaging to the
reputation of our company's products.

Reputable testing institutions have meticulously tested a variety of impact scenarios in order to
provide the public, the FAA, and the drone industry with supportable conclusions about risk.
ASSURE has set the standard for this work by releasing detailed reports with careful
documentation of their testing methodology and hundreds of pages of data. By contrast, the
limited information available about your demonstration prevents anyone from determining other
flaws in your methodology and conclusions.

Your video and blog post have been promoted in media around the world, yet nowhere in any of
your print or television appearances have you qualified the limited and unrealistic nature of your
test. As a safety researcher, surely you understand the detrimental impact on public perception
when purported scientific research is not presented with appropriate caveats and with an
opportunity for peer review and alternative views.

Unbalanced, agenda-driven research does substantial harm to our industry and to our company.
Policymakers at all levels of government have responded to sensational media coverage by
proposing and enacting new restrictions on drone ownership and use. These limitations prevent
people and businesses from using drones safely for beneficial purposes, such as performing
hazardous inspections or finding missing people. At least 195 people around the world have been
rescued from peril by drones, many of them saved by small drones such as DJI Phantoms. By
misleading the public and promoting fear about drones, you are undermining their benefits and
encouraging restrictions on their lifesaving uses.

We respectfully demand that you withdraw your research, remove the alarmist video from
circulation, and issue a corrective statement to the public and to all of the media outlets you have
appeared in, acknowledging that the configuration of the test was invalid given the flight
envelopes of the two aircraft tested, FAA testing standards, and the limited value of a single test.
Yours very truly,

Brendan M. Schulman
Vice President of Policy & Legal Affairs

cc:

Dr. Allan Crasto, Director, UDRI

Mary Ann Poirier Recker, Vice President and General Counsel, University of Dayton

Assessing the danger of drone strike: unique test bench to measure collision impact

by Birgit Bindnagel, Fraunhofer-Gesellschaft

No drones allowed in the vicinity of helicopters. Credit: Michael May

The rapid rise in the number of drones worldwide has been accompanied by increasing reports of
near misses with commercial aircraft. In 2017, while coming in to land, a Canadian passenger
aircraft actually collided with a drone, narrowly escaping catastrophe. Bird-strike tests for aircraft
are mandatory. To date, however, there is no equivalent standard test procedure for collisions
with drones. In order to to be able to fundamentally understand the consequences of a collision
between an aircraft and a drone, the Fraunhofer Institute for High-Speed Dynamics, Ernst-Mach-
Institut, EMI is now planning to build a test bench for recreating various collision scenarios with
complete drones.

The incidence of drones impacting air traffic has risen in recent years. In 2018, there were 158
cases reported at German airports. Indeed, Germany's federal police has warned of a massive
threat posed by unmanned aerial vehicles. This year, at the beginning of May, flight operations at
Frankfurt Airport had to be shut down completely for a short period following the sighting of
a drone. The total number of drones in private and commercial use in Germany is forecast to rise
to 847,000 by 2030, an increase of almost 80 percent.

Drones endanger not only aircraft coming into land but also low-flying helicopters. Pilots live in
fear of a drone hitting the cockpit windshield, the engine or the leading edge of the wings. Experts
are of the opinion that a collision with a drone would cause more damage to the aircraft than the
impact of a bird strike. Before being certified for use, aircraft must undergo a standard test to
assess their tolerance to bird strike. In the case of drones, however, there are no such regulations.
Researchers from Fraunhofer EMI in Freiburg are keen to see changes here. "From a mechanical
point of view, drones behave differently to birds and also weigh considerably more," explains Dr.
Sebastian Schopferer, one of the scientists working on this project. "It is therefore uncertain,
whether an aircraft that has been successfully tested against bird strike, would also survive a
collision with a drone."

A major threat to safety

Typical lithium-ion battery (weighing approximately 700 grams), as installed in a drone. Credit:
Fraunhofer EMI

Initial impact tests with drone batteries and motors have confirmed the danger. "Using
compressed air, we accelerated these two components to speeds ranging from 115 to 255 meters
per second and fired them at aluminum plates up to eight millimeters in thickness that were
mounted in a test bench," Schopferer explains. Given their weight, both the battery and the motor
can cause major damage: "There was substantial deformation and indentation of the plates, and
the drone battery and engine were completely destroyed." The outcome of the tests was recorded
with a high-speed video camera.

The primary objective of this series of tests is to determine the transfer of momentum at the
instant of impact with the two aforementioned components and to investigate the associated
damage to aircraft materials such as aluminum alloys and fiber composites. In parallel to these
dynamic investigations, researchers also conducted a number of quasi-static pressure tests in
order to determine the strength and rigidity of the drone components. These results will play an
essential role in the derivation of numerically efficient, predictive simulation models that
the aviation industry can then use to ascertain new and important findings about the impact
behavior of drones. Using such models during the design phase, it will be possible to assess the
resistance of new aircraft components to the impact of a drone.
Fragments of a drone battery following impact with an aluminum plate on the test bench. Credit:
Fraunhofer EMI

Acceleration tests with complete drones

In order to simulate realistic impact scenarios, researchers are now planning to construct a new
type of test bench for investigating the impact of complete drones with a maximum weight of
three kilograms and flying at speeds of up to 150 meters per second. "We will be able to
investigate the impact and fragmentation of complete drones during collision with both rigid and
flexible targets and thereby study the presumably catastrophic effects of a drone strike for an
aircraft," Schopferer explains.

"Tests in this weight class of drone have never been carried out before." The tests will be
conducted with a variety of drones, including both amateur and semiprofessional models,
weighing between one and three kilograms. In addition to aircraft manufacturers, these
investigations will also benefit aviation authorities, providing them with important information for
a more in-depth assessment of the danger to aircraft posed by drones.

Explore further
Report cites 241 near collisions between pilots, drones