Aerospace and Electronic Systems Magazine November 2016 Tutorial X - 84

Sense and Avoid for Unmanned Aircraft Systems
"Small" UAS are aircraft that will fly mostly in uncontrolled
airspace, where no (or very few) manned aircraft will be operating. FAA and other aviation authorities have not needed to exercise positive control in this environment, which includes very
low altitude Class G airspace, not near airports with controlled
airspace. Most manned flights involved in these airspaces have
been general aviation (GA) aircraft, helicopters (traffic, medical), or a variety of lower-end manned aircraft (gliders, weightshift controlled aircraft, gyroplanes, and balloons).
The fundamental issue for this diverse range of UAS types
is the need to consider how they will be integrated with (or kept
separate from) the current systems to manage manned aircraft.
The main driver so far for the integration of large UAS has been
the military interest in operating their UAS in civil airspaces (for
transit or mission purposes). The main interest in integration of
small UAS has been by the manufacturers, to be able to operate in
airspace that the FAA and other aviation authorities have had little
operational involvement with to date.

on a smaller scale, and are conducted outside of most manned aircraft operations. These airspaces are currently governed by VFR
regulations, though the UAS operations will vary widely by mission, many of which will require operating more closely to people
and obstacles (for photography, inspections, etc.) than current aviation separation standards permit. This is a new and rapidly evolving area, which will require new separation definitions and UAS
airworthiness requirements, to permit them to operate safely and
predictably in these environments.

II. TAXONOMY OF SENSE AND AVOID
The process of SAA consists of a series of discrete steps that begins with sensing the environment around the UA and ends with
taking an action to avoid a fully determined threat to the UA. For
this tutorial, we dissect the SAA process into three fundamental
tasks:
1. Sense - methods for surveilling the environment around the
aircraft

B. WHAT IS SENSE AND AVOID?
To address the lack of "see and avoid," the most fundamental
method would be to provide the UAS pilot with equivalent capabilities, such as a "first-person view" with onboard cameras
transmitted to the ground station. Apart from data link limitations,
most of these solutions have proven to be very limited, with either
a constrained field of view (FOV) ("like looking through a soda
straw"), or a limited ability to discern a target on a video picture, due to resolution or contrast concerns. Beyond this immediate solution, SAA consists of methods for the UAS to "sense and
avoid" other aircraft or obstacles, either onboard the UA or with
the ground station in the loop.
There will be fundamental differences between SAA for
large UAS (operating in controlled airspace) and small UAS
(operating at low altitudes in minimally or uncontrolled airspace). Large UAS will be required to deal with manned aircraft
operating in the same airspace, as the main obstacle to avoid.
To be integrated, these UAS will need to interact at the levels
of separation defined for these airspaces (en route and terminal), operating under appropriate flight rules [usually instrument flight rules (IFR) or visual flight rules (VFR)]. The SAA
capability will have to react with avoidance maneuvers that are
appropriate and expected, to permit the UAS to fly with equivalent levels of safety to manned aircraft. Currently, since large
UAS do not have a sufficient SAA solution, the alternative is to
permit UAS to fly under a waiver of current flight rules, which
usually means that the UAS operation must be segregated from
the manned operations to ensure safety. If SAA solutions are
not able to be certified, segregated airspace or alternative flight
rules must be used for the UAS when it is operating in this environment. Therefore, it is the goal for SAA to be developed sufficiently to permit large UAS to interact with manned aircraft in
an expected and timely fashion.
For small UAS operating at low altitudes in uncontrolled airspace, the operating environment is much different-operations are
84

2. Detect - analysis to determine if there are aircraft or obstacles
in that environment, and to evaluate if they are, or will be, a
threat to the UA
3. Avoid - evaluation of the actions that the UA should take to
reduce or remove the threat of the detected aircraft or obstacle.
These three fundamental tasks of an SAA system can be implemented in different ways, giving rise to several architectural and
technical solutions. These solutions can then be classified using
different taxonomies. Some of these classifications are introduced
in this section, while the different segments are focused on in the
subsequent part of the paper.
A general approach for classification that involves all
three parts is based on the physical location of information
sources and processing/decision making centers. In fact,
these elements can be based onboard the UA, or be located
on the ground. The first implementation is defined as onboard
SAA, while the second can be named as ground-based or offboard SAA. Moreover, a mixed architecture can be used [4].
For example, information sources can comprise both ground
and airborne systems. Also, some UAS SAA systems can be
designed to work on the basis of remote pilot commands,
leaving autonomous avoidance as only the last-moment safety resource. This range of approaches can be summarized in
the following three scenarios, though there are other combinations of sensor technologies and detection and avoidance
processing.
1. Sensing takes place on the UA. Detection and tracking processing of target aircraft may take place on the aircraft or in
the GCS. Processed sensor information is transmitted from
the UA to the GCS, where it is evaluated by the pilot, who
(with the help of GCS processing) decides if an action needs
to be taken to avoid a target aircraft and what that action is.
The pilot transmits control actions up to the UA by means of
a line-of-sight (LOS) or beyond line-of-sight (BLOS) data

IEEE A&E SYSTEMS MAGAZINE

NOVEMBER 2016, Part II of II

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine November 2016 Tutorial X