Aerospace and Electronic Systems Magazine September 2016 - 60

Proof-of-Concept Airborne Sense and Avoid System

Figure 8.

HMD filter impact on CPDS conflict probe. Left: Conflict probe without HMD filter. Right: Conflict probe with HMD filter.

tion of new features as UAS, SAA, MOPS, and the quantitative
definition of "well clear" evolve.
The SS flight test investigated the impact of a horizontal missdistance (HMD) filter on CPDS. The HMD filter is additional
logic to reduce the false alarm from the well clear τ requirement
by using measured intruder acceleration to detect HMDs that
are sufficiently large so as not to be a threat, similar to those in
commercial TCAS II systems (version 7.0 or later) [10]. Figure
8 illustrates the impact of HMDs on the CPDS conflict probes,
notice that without the HMD filter there is additional padding
around the areas that projects a well clear violation. The HMD
filter can significantly influence the conflict probe geometry [11],
and ultimately the pilot's decision to execute SS maneuvers. For
the encounter scenario illustrated in Figure 8, the pilot decided to
maneuver based on conflict probe generated without the HMD
filter. With the HMD filter, the pilot decided to maintain its path
because the filter removed the "false alarm" conflict probe areas
on ownship's heading, providing him confidence that well clear
will not be breached. The HMD filter requirement on SS display
is preferred from an ATC standpoint, as it reduces the frequency of
UAS maneuvers, keeping aircraft traffic trajectories predictable,
minimizes flight plan deviations, and makes it easier for controllers to manage.

AUTORESOLVER
Autoresolver is an SS algorithm developed by NASA Ames implemented into the Vigilant Spirit Control Station (VSCS), a research
pilot/operator interface to control UAS managed by the Air Force
Research Laboratory (AFRL) [12]. Originally designed for ATC
ground-based separation assurance system to manage en route aircraft, the Autoresolver algorithm was adapted to assist UAS pilots
60

in remaining well clear [13]. Unlike CPDS, VSCS/Autoresolver is
a directive guidance system which issues SS maneuver advisories
that direct the UAS pilot to fly a specific heading or altitude in
order to remain well clear.
Autoresolver calculates ownship and intruder trajectories each
time surveillance data is updated. It analyzes the ownship trajectory against intruder trajectories in order to calculate CPA, detect
any upcoming well clear violations and determine the appropriate
alert/maneuver to resolve the potential conflict. Autoresolver's
performance is particularly sensitive to the ownship state and
surveillance data quality used for trajectory synthesis because its
logic reevaluates threat detection and resolution at each surveillance data update instance. Ideally, Autoresolver would detect
and resolve potential well clear violations with a single attempt.
However, during the flight test Autoresolver issued intermittent
alerting and inconsistent resolution maneuvers due to the variation in horizontal and vertical CPA prediction errors, shown in
Figure 9 for the encounter scenarios flown. Data points under
each scenario bins are CPA prediction errors calculated by Autoresolver throughout the encounter. The pilots maneuvered away
from the intruder aircraft during the encounter based on Autoresolver warnings, so a "truth" CPA was not available. Instead, the
Autoresolver-predicted CPA was compared with an expected CPA
based on the flight test cards. Horizontal CPA error is the difference between the predicted horizontal CPA and the expected horizontal CPA specified for each scenario. A negative error means
that the pilot was on a trajectory that would have been closer than
expected if no SS maneuver was executed. Corrected vertical CPA
is the difference between the predicted vertical CPA and the actual altitude difference. This correction factors out actual flight
altitude differences such that both aircraft appear to be coaltitude.
A box-whisker plot format summarizes CPA prediction error sta-

IEEE A&E SYSTEMS MAGAZINE

SEPTEMBER 2016

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine September 2016