Research On 4D Trajectory Planning Algorithm Based On Multi-source Data Fusion Method

With the high-speed development of global aviation industry, there have been rapid increases in the shortage of airspace resources and air traffic flow congestion. To resolve effectively the problem of airspace management under the condition of high density, small separation, and great traffic flow,conflict-free 4D trajectory prediction and planning becomes one of the core technologies of next new generation ATC automation system. Therefore, in the study, in order to achieve the core technology, a conflict-free 4D trajectory prediction and planning algorithm based on multi-source data fusion is proposed to predict changes of aircraft flight status in advance, as well as providing the necessary information so controllers can handle aircraft conflicts timely. The paper focused on the key issues including a generating algorithm of initial nominal flight profiles, a hybrid model combining airspeed fitting with meteorological revision model, as well as a conflict-free 4D trajectory planning method.Firstly, we propose a generating algorithm of nominal flight profiles combining improved K-means algorithm with BADA model that effectively improves the accuracy of them. On the one hand, according to the characteristics of the trajectory data, a SWED algorithm is proposed to measure the similarity between different altitude profiles that break the constraints of the traditional K-means algorithm. On the other hand, in order to extract a more accurate speed profile, the TAS determined by speed triangle relationships, between ground-speed and wind-speed obtained via IDW interpolation,and the CAS determined by BADA model are used to describe the airspeed profiles above and below the airspeed transition altitude, respectively.Secondly, for the problem of how to improve the accuracy of trajectory estimation, we propose an airspeed fitting algorithm based on a improved method of least squares to determine status information of arbitrary trajectory point, and an en-route meteorological revision model through using Gressman interpolation to combine GRIB with AMDAR data.Thirdly, for potential conflicts problem existing in 4D trajectories predicted, we put forward a conflict-free 4D trajectory planning algorithm, which combines 4D space-time grid and space geometry model to implement conflict detection with dynamic grouping strategy to achieve planning through departure time adjustment, to reduce conflict rate in advance and relieve workload of controllers.Finally, we set up a simulation system platform of conflict-free 4D trajectory prediction via World Wind, summarize and present the research content and results achieved of this paper.