Aerial Threat Detection around Critical Infrastructures: Drone Trajectory Analysis and Behavior Profiling

In recent times, the utilization of drones has seen a significant increase across both civilian and military domains. Within the military sphere, these UAVs (Unmanned Aerial Vehicles) exhibit the capacity to engage targets at extended distances, disrupt critical infrastructure, and conduct comprehensive facility assessments via ISR (Inspection, Surveillance, and Reconnaissance) operations. To prevent such events from happening, it is required a calibrated system to detect drones and conduct an analysis of their intentions based on their movement and behavior, while also considering the economic implications of deploying costly missiles for every drone interception. This is the principal focus of this thesis: after detailed research, some credible trajectories that could be followed by attacking drones were simulated and fed to a classifying system with the scope to recognize which patterns were attacks and which were harmless. The classifier is also supported by a trajectory-forecasting system, that can help direct the decision.