Visual demonstrations of performance characteristics of surface surveillance

In this paper we describe the use of visual demonstrations (both still images and videos) to aid air traffic control subject matter experts (SMEs) in estimating the required performance characteristics for a surface surveillance system. The demonstrations focused on performance characteristics of lo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Stanley, R. M., Colavito, A., Rammelsberg, K., Stelzer, E. K., Tuomey, D.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this paper we describe the use of visual demonstrations (both still images and videos) to aid air traffic control subject matter experts (SMEs) in estimating the required performance characteristics for a surface surveillance system. The demonstrations focused on performance characteristics of location accuracy, orientation accuracy, latency, and update rate. The demonstrations were aimed to increase validity of performance characteristic estimates by allowing SMEs to visualize the consequences in an airport surface context. Still images were created to show the impact of surveillance error for aircraft location and orientation by depicting an opaque aircraft location on a satellite image and using ghost images to depict error. Videos of aircraft moving on and around the airport surface were created to show how surveillance with a specified latency or reduced update rate would appear. These demonstrations influenced the value and variability of performance characteristic estimates, relative to initial discussions without visual demonstrations. The visual demonstrations also increased the consensus and confidence that the team had in the estimates. The distinct advantage of the visual demonstrations is that they elicited performance characteristics estimates that were more valid than simple verbal discussions of values, while being less expensive and faster than full human-in-the-loop simulations.
ISSN:2155-7195
2155-7209
DOI:10.1109/DASC.2012.6382397