Abstract: Recent military operations in urban environments are changing the requirements imposed on sensing technologies. The final goal remains threat mapping within the area of operation, but the environmental constraints and the intrinsic nature of urban threats are radically novel. AUDIS (Acoustic Urban Threat Detector for Improved Surveillance Capabilities) consists of a novel cognitive sensor that offers flexibility and adaptivity to the encountered scenarios, while also ensuring an improvement in recognition and characterization of such ground threats. AUDIS specifically aims to increase the state-of-the-art capabilities in threat detection, localization, classification and identification, supporting urban situational awareness. To achieve this goal, the novelty also resides in the sensor concept and proposed logical architecture. AUDIS exploits the conceived sparse/arrayed antenna configuration by means of an innovative ensemble of digital processing stages that support a learning-based, fully adaptive approach to urban threat recognition and characterization. The “knowledge” on the scenario and the expected/actual threats will be collected, stored and managed, exploited and grown by AUDIS to form the “base” on which the expected capability improvements can be found.