Automated processing and precise trajectory analysis of airport surface surveillance data based on multiview discriminant feature selection

In the automatic processing of airport surface surveillance data and accurate trajectory analysis, DRO (Distributed Robust Optimization) distributed robust optimization methods are often used to construct automated data processing models, but they are susceptible to the embedding complementarity problem, resulting in poor consistency indicators. Therefore, a method for automated processing and precise trajectory analysis of airport surface surveillance data based on multi-view discriminative feature selection is proposed. Specifically, coupled matrix decomposition is used to jointly represent multi-view compatible information. Nuclear norm regularization is incorporated into a joint feature selection framework. Embedding complementarity processing is performed considering the multi-view discriminative scene relationship, determining the data's latent representation state and applying a regularization parameter. A Gaussian kernel function is used to calculate sample similarity weights, adjusting the linear correlation of data processing, and constructing an automated surveillance data processing model. The coordinates of the far-point reference point are determined based on the number of radar track points, and terminal airspace is partitioned. Adjustments are made based on the crossing status of the radar track to determine the two-dimensional coordinates of the turning point. Hierarchical clustering and track set classification are used to generate an automated data processing and precise trajectory analysis algorithm, completing the automated processing and precise trajectory analysis of airport surface surveillance data. Experimental results show that the proposed method achieves consistently high Consistency indicators under different trajectories, without exhibiting abnormal polarization issues. This demonstrates its robustness and good processing and analysis performance