Combined Radar and Camera Drone Detection in Urban Environment: A Simulation-based Approach

Unmanned Airborne Systems (UAS) have gained popularity in recent years. Drone pilots sometimes operate in restricted areas where they can involuntarily disrupt human activities, they sometimes deliberately conduct illicit activities, or some can weaponize their UAS. A significant challenge associated with counter-UAS is the disproportionate cost difference between the detection/mitigation systems and customer-grade UASs. In this paper, we focus on the cost-efficient detection of UAS activities in urban environments. More specifically, we present a simulation platform designed to study the concurrent use of AI-powered camera systems and radar. Those AI-powered camera systems can be sold as software stacks that are supposed to be camera-agnostic. The objective of our simulation approach is to ease the selection of camera models, lenses, and the positioning of the cameras in order to complement radar coverage.