IEEE Consumer Electronics Magazine - January/February 2023 - 64

Increasing Drone Safety Through Illumination
Evaluation and Testing
To evaluate the feasibility of our system
architecture, we conducted two different tests
measuring two different aspects, power consumption,
and the effect of our application-oriented
unit on real-world scenarios.
Power Consumption Test As the space for
power sources on a consumer-level drone is
limited, it was important to us to determine
how much the battery that supported our application
unit would last. We were interested to see
how the potential colors and lighting patterns
that could be used on a consumer drone affected
the overall power consumption. As such, we included
colors that are commonly used in aviation,
such as red, green, white, and blue, and patterns,
such as solid, flashing, and revolving.
Our results found that the lighting pattern
affected the battery life significantly more than
the type of color. To give a frame of reference,
we initially tested the total life of the battery
pack with nothing enabled, which was 8.4 hours.
The solid configuration in comparison averaged a
1.76 hours for all four colors. Among the four, white
was the longest lasting with 1.81 and the shortest
red, with 1.72. In terms of power consumption, the
solid pattern consumed almost five times* the
power compared to the stand-by mode with an
average current of 1138 mA compared to 238 mA.
The second pattern that consumed themost power
was the revolving colors pattern with an average
current of 913 and a duration of 2.19 hours. The
least power-consuming pattern was the flashing
pattern, with an average of 5.29 hours and 378 mA
of overall average current. White and red were the
lowest and highest power-consuming color types
with 5.33 and 5.25 hours, respectively, mirroring
the results in the solid pattern.
Human-in-the-Loop Assessment The perception
of the drone by both drone operators and
the people that will be in the vicinity of the drone
is critical if drones are to be used in cities and
other populated areas. As such, it is important to
test the system in the real-world environment that
it will be deployed in. It is of specific importance
to determine how the illumination system would
affect the interactions humans have with drones.
We chose a task that is very relevant to human
64
Figure 6. Experiment setup for the human-in-theloop
assessment.
drone interaction (HDI): determining the distance
of the drone froma specific vantage point.
The experiment was held in an outdoor setting.
The experiment setup is depicted in Figure 6. The
protocol was evaluated and approved by theUniversity
of Florida IRB board. The vantage point the participants
stood made up the center of a circular
layout with predetermined points for the drone to
hover. There were four such drone hovering points,
each located at the cardinal directions. The points
on the east and west sides were 5 meter away, while
the points on the north and the south were 6meter
away, respectively. The two drone configurations
were unilluminated and illuminated with white. In
total, the participants made four guesses and the
accuracy was determined by normalizing the actual
distance and the distance participants stated. For
this task, the illuminated condition resulted in an
error rate of 14.77%, whereas the unilluminated conditionhad
IEEE Consumer Electronics Magazine - January/February 2023

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2023
