Evaluation of multi-drone sound source tracking algorithms

This study reviews and evaluates methods for sound source tracking using multiple microphone array-equipped drones. These previous studies performed multi-drone sound source tracking by the MUltiple SIgnal Classification (MUSIC) algorithm with various ways to triangulate localisation outputs from each drone. Recently, techniques to reduce the influence of drone rotor noise have also been explored, such as using a rotor noise reduction Wiener filter and noise correlation matrix estimation. In addition, methods to navigate the drones to optimise microphone array placement have also been demonstrated. Here, we evaluate the performance of these methods and demonstrate whether improvements can be made by combining components proposed in recent studies.