(Created page with " == Abstract == traffic systems have long relied on automated short-term conflict prediction algorithms to warn controllers of impending conflicts (losses of separation). Th...") |
m (Scipediacontent moved page Draft Content 285210642 to Robinson et al 2010a) |
(No difference)
|
traffic systems have long relied on automated short-term conflict prediction algorithms to warn controllers of impending conflicts (losses of separation). The complexity of terminal airspace has proven difficult for such systems as it often leads to excessive false alerts. Thus, the legacy system, called Conflict Alert, which provides short-term alerts in both en-route and terminal airspace currently, is often inhibited or degraded in areas where frequent false alerts occur, even though the alerts are provided only when an aircraft is in dangerous proximity of other aircraft. This research investigates how a minimal level of flight intent information may be used to improve short-term conflict detection in terminal airspace such that it can be used by the controller to maintain legal aircraft separation. The flight intent information includes a site-specific nominal arrival route and inferred altitude clearances in addition to the flight plan that includes the RNAV (Area Navigation) departure route. A new tactical conflict detection algorithm is proposed, which uses a single analytic trajectory, determined by the flight intent and the current state information of the aircraft, and includes a complex set of current, dynamic separation standards for terminal airspace to define losses of separation. The new algorithm is compared with an algorithm that imitates a known en-route algorithm and another that imitates Conflict Alert by analysis of false-alert rate and alert lead time with recent real-world data of arrival and departure operations and a large set of operational error cases from Dallas/Fort Worth TRACON (Terminal Radar Approach Control). The new algorithm yielded a false-alert rate of two per hour and an average alert lead time of 38 seconds.
The different versions of the original document can be found in:
DOIS: 10.2514/6.2010-9294 10.2514/1.51898
Published on 01/01/2010
Volume 2010, 2010
DOI: 10.2514/6.2010-9294
Licence: CC BY-NC-SA license
Are you one of the authors of this document?