Towards mobile microphone array based UAV tracking

The emergence of new technologies for UAVs and their commercial availability offer great opportunities in supporting humans in a wide variety of tasks. However, when used as a means of attack, single UAVs or swarms may create a potential asymmetric threat situation, which cannot be satisfactorily countered with existing sensor technology. Currently, multimodal approaches are being investigated in which UAVs can be detected, localized, and tracked using a composite of different sensors. In the development of such multisensor solutions, the acoustic domain has emerged as an indispensable element. However, existing systems relying on acoustic sensors primarily focus on stationary operation and detection of a single UAV. In this work, we present approaches for the development of a mobile acoustic sensor system that can detect and track multiple drones based on the direction of sound incidence. In contrast to the majority of solutions existing in the literature, where the focus lies on either detection or localization, we propose joint estimation of UAV sound presence and incidence. We evaluate our proposed system on a dataset acquired in the course of a measurement campaign conducted at a military camp near Bure, Switzerland, for a stationary as well as for a mobile scenario, where a microphone array was mounted on the roof of a vehicle.