Aviation Surveillance Information Fusion Based on Ensemble Learning

At present, with the comprehensive development of economy and technology in our country, the air transportation industry has also ushered in a golden period of development. The air traffic volume increases year by year, and the air route traffic volume will increase, which will lead to the need to carry more aircraft on the limited channel, resulting in the congestion of the channel and the potential safety problems. In order to ensure that the aircraft can fly safely, it is necessary for traffic management personnel to maintain the order of the aircraft in the channel according to the aviation monitoring information. Therefore, the accuracy of aviation surveillance information is particularly important. As a traditional track fusion algorithm, the Kalman filtering has the problem of requiring accurate error estimation, insensitivity to noise, and long calculation time in the case of large data volume. In this paper, a method of air surveillance information fusion based on ensemble learning is proposed, which can predict and fuse multiple air surveillance sources, i.e. multiple radar surveillance information, so as to obtain more accurate position estimation of the monitored target and reduce the error caused by the accuracy of radar itself, geographical location and surrounding environment.