The main objective of this project is to research and design new Reinforcement Learning, Deep Learning and Distributed Logging Technologies for the construction of a modular platform for the prediction, simulation and detection of threats in Remotely Piloted Aircraft Systems (RPAS), as well as the generation of flight and mission models for the progressive and adaptive training of pilots.