A study of 4D trajectory prediction based on machine deep learning

Trajectory prediction is the core module of modern air traffic management system. Focused on the trajectory prediction model and key algorithm of the system, this paper tries to overcome the disadvantages of the traditional linear prediction method via employing machine learning technology in the problem. The statistical historical flight data of different aircraft type is taken as the training sample to build the predictive model. Besides, current state information is used as the input data of the model and to get the predictive flight trajectory information. Then machine learning technology is adopted to improve the efficiency. The result shows that the proposed method has obviously higher performance compared to the traditional predictive methods.