assumption for this conclusion is that the UAS is presumed to be in a probable failure mode (such
as a loss of power) and to fall down, impacting a person. Risk posed to people on the ground by
lateral impacts, such as a forward-speed loss-of-control scenario, are already addressed by proposed
part 107. The ARC did not attempt to quantify the current risk of such a failure mode, or an
acceptable failure rate, or to specify the acceptable probability of a human impact occurrence. On
guidance from the FAA, the ARC adopted a conservative assumption that any UAS flown over
people may experience a failure. Therefore, the ARC’s recommendations focus on the severity of
injury that is acceptable assuming the UAS makes impact with a person.
Based on the information received, the ARC agreed that the metric used to quantify an acceptable
probability of an AIS level 3 and above injury should be the impact energy of the small UAS,
expressed in joules (J)/centimeter² (cm²). For a particular model of small UAS to qualify for
operations over people, the manufacturer of that model will therefore have to certify that the
product’s impact energy, as measured by a test established by an industry consensus standards body,
does not, in the most probable failure modes, exceed a specified threshold. The intent of the test
should be to establish the typical or likely impact energy of the most probable failure mode, and not
simply the worst case condition.
4.1.1
Category 1 Performance Standards
The ARC believes that there are small UAS that pose a level of risk that is so low that they are
relatively safe to operate over people without being subjected to regulation beyond proposed part
107. As explained below, for this category of UAS, the ARC recommends a maximum risk impact
threshold of a 1% chance of AIS level 3 or greater injury, based on kinetic energy transfer at impact.
The impact kinetic energy transfer standard and industry consensus standards are more fully
explained in the Category 2 discussion. For simplicity, the ARC recommends using a weight-based
measure instead of an impact kinetic energy measure for Category 1.
The ARC was presented with several studies and analytical methods that could be used to assess the
risk of a weight-based category of small UAS allowed to operate directly over people. The
presentation from Dr. Paul Wilde of the FAA discussed methods used for assessing individual and
collective public risk for commercial space launches. He provided one example, subject to certain
assumptions, where a UAS weighing 0.55 lbs. (250 grams) operated over people had a probability of
serious injury or fatality consistent with existing levels of safety for non-participating people when
exposed to aviation risks. The presentation from Dr. David Arterburn from ASSURE’s University
of Alabama, Huntsville correlated various human injury thresholds with risks associated with
sporting events where people are exposed to risks from fast moving balls and other objects that
could cause an injury or, in extreme cases, a fatality, yet the public is generally accepting of those
risks.
In addition, Dr. Arterburn’s study showed results that suggest the kinetic energy-based injury
calculation was overstated as it relates to UAS because of the differences between a simple ball and
actual UAS design. Specifically, UAS were found by ASSURE to deflect and tumble on impact,
resulting in only an average 38% kinetic energy transfer during an impact, and therefore a
significantly lower expected severity of injury compared to shrapnel impacts that were the basis of
those prior estimates of harm, taken from studies of explosives and ordinance effectiveness.
April 1, 2016
Page 6