Detection of convective clouds using meteorological data fusion for aviation safety support

With the growth of the aeronautical industry worldwide, it is important task for the next generation of air traffic control systems generating technologies that reduce risks and maintain a constant monitoring of atmospheric meteorological phenomena. Many meteorological phenomena generate risks and delays in air operations. An objective here is to provide an overview of the main meteorological processes that affect flights seen from the perspective of control centers. This work presents the development of a methodology that allows the automated integration of heterogeneous and asynchronous meteorological data, in order to determine the location and characteristics of meteorological phenomena and to analyze their behavior in order to detect those atmospheric formations that can put in risk air operations. For this purpose, artificial intelligence systems are used to extract relevant information from several sources such as: meteorological satellite images in different spectra; infrared, visual and water vapor; Icing models and data from the meteorological gardens of all aerodromes in Colombia. With all this information, the characteristics of the sources are evaluated in real time looking for meteorological formations that could be considered as relevant and then they are plotted in a georeferenced and organized way on a map. An interference matrix is obtained, which summarizes the data obtained from the individual analyzes of each source in order to obtain the geolocation of vertical formation of clouds, classifying and labeling automatically each formation providing support information to experts and other users of the aeronautical ecosystem to support decision-making.