| With the rapid economic development,people’s demand for civil aviation is increasing.The scale of the fleet and air routes has expanded,the airspace resources are in short supply,and air route congestion and conflicts frequently occur.The new generation of air traffic control automation system proposes to take the trajectory-based operation mode as the core technology of the next stage.Precise four-dimensional trajectory prediction can improve the control efficiency of air traffic control and timely provide the best means of conflict resolution.The air environment is complex and changeable,and factors that affect the accuracy of trajectory prediction include: aircraft performance,airspace environment,meteorological factors,flight intentions,etc.The mainstream trajectory prediction only considers one or two influencing factors,and the prediction accuracy needs to be improved.The research idea of the paper is to conduct initial trajectory prediction based on historical flight data mining and analysis,construct aircraft motion model,describe flight operation status,and modify the trajectory prediction results with updated aircraft status information to improve the prediction effect.Firstly,it summarizes the research status and methods of trajectory prediction.The historical flight data is preprocessed,the historical flight trajectory cluster analysis based on the hierarchical clustering algorithm is established,the trajectory of the cluster center is extracted,and the initial trajectory prediction is realized through the decision tree algorithm.Then the aircraft position and speed are described by data fitting,the aircraft motion model is constructed,and the initial trajectory and past time prediction results are corrected by updating the flight information of the aircraft.Finally,a real flight from Capital Airport to Hongqiao Airport was used for instance verification.The aircraft motion model constructed can describe the flight status well and realize the initial flight path prediction.Using the model combined with the updated radar track data can control the flight overtime error within 6s within a 100 km voyage and can correct the initial track position prediction.Figure 27;Table 10;Reference 53... |
