Due Regard Encounter Model Version 1.0. Griffith, J. D., Edwards, M. W., Miraflor, R. M., & Weinert, A. Technical Report DTIC Document, 2013.
Due Regard Encounter Model Version 1.0 [link]Paper  abstract   bibtex   
Airborne safety critical systems must undergo extensive validation under realistic conditions prior to certification in the National Airspace System (NAS). Collision avoidance systems for manned aircraft and sense and avoid (SAA) systems for Unmanned Aircraft Systems (UASs) provide a safety critical function. They ensure separation when other safety layers have failed to maintain separation (e.g., the underlying airspace structure, airspace procedures, and air traffic control). These complex systems are assessed in realistic large scale Monte Carlo simulations and flight tests to prove that they meet the desired levels of safety. Fundamental to these simulations is the use of realistic encounter situations between aircraft which are defined by the relative geometry and behavior of the aircraft during the encounter. The geometry and dynamic behavior of aircraft during encounters are captured in encounter models. Encounter models provide a statistically sufficient set of features which are estimated from a large collection of observed encounter events. The initial and continuing evaluation of the Traffic Alert and Collision Avoidance System (TCAS) for manned aircraft illustrates the necessity of encounter models to estimate system effectiveness in a wide variety of encounter geometries [1–3]. Lincoln Laboratory has previously developed encounter models for the NAS. These models were built from radar data collected from radars across the NAS [4]. There are fundamentally two types of encounters: correlated and uncorrelated. In the first type, at least one aircraft in the encounter is receiving air traffic control (ATC) services, and both aircraft are transponder equipped. In the second type, ATC services would no longer be provided because either one aircraft is not transponder equipped or both aircraft are flying under visual flight rules (VFR). The first type is termed “correlated” because there is active coordination provided by ATC prior to the loss of separation. It is therefore critical that the model capture this coordination between aircraft prior to the loss of separation. In the second type of encounter, a lack of ATC services results in uncoordinated loss of separation. Hence, these encounters are termed “uncorrelated”—i.e., aircraft blunder into one another. This uncorrelated feature is exploited by modeling each aircraft individually and then simulating the uncoordinated loss of separation. This report describes the development of an encounter model that can be used to evaluate SAA systems operating on UAS flying due regard in oceanic airspace. An aircraft flying due regard is not operating under International Civil Aviation Organization (ICAO) flight procedures and is responsible for maintaining its own separation from other aircraft [5]. More specifically, a due regard encounter model describes the statistical distribution of close encounter situations that are expected to occur in oceanic airspace when an UAS is operating due regard. A due regard encounter model is an uncorrelated model because ATC intervention is unlikely. The next section gives an overview of all the encounter models that Lincoln has developed and describes when it is appropriate to use each one. The following section then gives an overview of this report.
@techreport{griffith_due_2013,
	title = {Due {Regard} {Encounter} {Model} {Version} 1.0},
	url = {http://oai.dtic.mil/oai/oai?verb=getRecord&metadataPrefix=html&identifier=ADA589692},
	abstract = {Airborne safety critical systems must undergo extensive validation under realistic conditions prior
to certification in the National Airspace System (NAS). Collision avoidance systems for manned
aircraft and sense and avoid (SAA) systems for Unmanned Aircraft Systems (UASs) provide a
safety critical function. They ensure separation when other safety layers have failed to maintain
separation (e.g., the underlying airspace structure, airspace procedures, and air traffic control).
These complex systems are assessed in realistic large scale Monte Carlo simulations and flight tests
to prove that they meet the desired levels of safety. Fundamental to these simulations is the use
of realistic encounter situations between aircraft which are defined by the relative geometry and
behavior of the aircraft during the encounter. The geometry and dynamic behavior of aircraft
during encounters are captured in encounter models. Encounter models provide a statistically
sufficient set of features which are estimated from a large collection of observed encounter events.
The initial and continuing evaluation of the Traffic Alert and Collision Avoidance System (TCAS)
for manned aircraft illustrates the necessity of encounter models to estimate system effectiveness
in a wide variety of encounter geometries [1–3].
Lincoln Laboratory has previously developed encounter models for the NAS. These models were
built from radar data collected from radars across the NAS [4]. There are fundamentally two types
of encounters: correlated and uncorrelated. In the first type, at least one aircraft in the encounter
is receiving air traffic control (ATC) services, and both aircraft are transponder equipped. In the
second type, ATC services would no longer be provided because either one aircraft is not transponder
equipped or both aircraft are flying under visual flight rules (VFR). The first type is termed
“correlated” because there is active coordination provided by ATC prior to the loss of separation.
It is therefore critical that the model capture this coordination between aircraft prior to the loss of
separation. In the second type of encounter, a lack of ATC services results in uncoordinated loss
of separation. Hence, these encounters are termed “uncorrelated”—i.e., aircraft blunder into one
another. This uncorrelated feature is exploited by modeling each aircraft individually and then
simulating the uncoordinated loss of separation.
This report describes the development of an encounter model that can be used to evaluate SAA
systems operating on UAS flying due regard in oceanic airspace. An aircraft flying due regard
is not operating under International Civil Aviation Organization (ICAO) flight procedures and is
responsible for maintaining its own separation from other aircraft [5]. More specifically, a due regard
encounter model describes the statistical distribution of close encounter situations that are expected
to occur in oceanic airspace when an UAS is operating due regard. A due regard encounter model
is an uncorrelated model because ATC intervention is unlikely. The next section gives an overview
of all the encounter models that Lincoln has developed and describes when it is appropriate to use
each one. The following section then gives an overview of this report.},
	urldate = {2015-04-06},
	institution = {DTIC Document},
	author = {Griffith, J. D. and Edwards, Matthew W. and Miraflor, Raymond M. and Weinert, Andrew},
	year = {2013},
}

Downloads: 0