Probabilistic conflict detection for commercial aircraft near airports

Pienaar, Leanne Jane (2015-03)

Thesis (MEng)--Stellenbosch University, 2015.

Thesis

ENGLISH ABSTRACT: Increasing air traffic and urbanisation has led to a cluttered airspace, particularly near airports, where both complex terrain and multiple moving obstacles are frequent. Accurately and efficiently predicting violations in safe separation criteria for commercial aircraft, a process called conflict detection, is therefore crucial in assessing risk associated with threats of collision. Existing avoidance systems in operation such as TCAS, EGPWS and ATC exhibit shortcomings, leaving room for uncertainty and possible conflict scenarios. A single on-board system capable of minimising errors in prediction would inform conflict resolution decisions more accurately as well as support the notion of free flight, an objective of next-generation air traffic management systems. This thesis investigates the viability of a modern algorithm, probability flow, as a method of probabilistic conflict detection for commercial aircraft in airport environments. Simulation results for realistic flight scenarios are presented in comparison with a ground-truth result obtained through Monte Carlo simulation. Observations are made regarding the suitability of probability flow for real-world application. It is found that probability flow is capable of calculating a tight upper bound to the probability of conflict quickly and accurately for most conflict scenarios. However, unreasonably large overestimates on the probability of conflict are obtained when flying parallel to an obstacle conflict region. This problem could lead to a high frequency of false alerts, particularly in aborted landing scenarios and at airports operating parallel runways. It is therefore advised that further research be conducted to resolve this problem before probability flow can be reliably implemented in an airport environment.

AFRIKAANSE OPSOMMING: Toenemende lugverkeer en verstedeliking het gelei tot ‘n deurmekaar lugruim, veral naby lughawens, waar beide komplekse terrein en verskeie bewegende struikelblokke gereeld voorkom. Akkuraat en doeltreffende voorspelling van oortredings in veilige skeidingskriteria vir kommersiële vliegtuie, naamlik konflik opsporing, is dus van kardinale belang in die beoordeling van die risiko wat verband hou met dreigemente van ‘n botsing. Bestaande vermyding stelsels in werking soos TCAS, EGPWS en ATC toon tekortkominge, wat ruimte laat vir onsekerheid en moontlike konflik scenario’s. ‘n Enkele aanboordstelsel, wat in staat is om foute in voorspelling te verminder, sou konflikresolusie besluite meer akkuraat in kennis stel, asook om die idee van vrye vlug te ondersteun, ‘n doelwit van toekomstige lugverkeer beheerstelsels. Hierdie tesis ondersoek die lewensvatbaarheid van ‘n moderne algoritme, waarskynlikheidsvloei, as ‘n metode van probabilistiese konflik opsporing vir kommersiële vliegtuie in die lughawens omgewing. Simulasie resultate vir realistiese vlug scenario’s word aangebied in vergelyking met ‘n grond-waarheid resultaat wat verkry word deur middel van Monte Carlo simulasie. Waarnemings word gemaak ten opsigte van die geskiktheid van waarskynlikheidsvloei vir die werklikheid. Dit is bevind dat waarskynlikheidsvloei in staat is om die berekening van ‘n stywe bogrens tot die waarskynlikheid van konflik vinnig en akkuraat te bepaal vir die meeste konflik scenario’s. Tog is daar ‘n onredelike groot oorskatting op die waarskynlikheid van konflik wat verkry word wanneer ‘n vliegtuig parallel met ‘n hindernis konflik streek vlieg. Hierdie probleem kan lei tot ‘n hoë frekwensie van valse waarskuwings, veral in mislukte landing scenario’s en by lughawens wat van parallel aanloopbane gebruik maak. Dit word dus aanbeveel dat verdere navorsing gedoen word om die probleem op te los voordat waarskynlikheidsvloei betroubaar in ’n lughawe omgewing geïmplementeer word.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/97033
This item appears in the following collections: