Acoustic simultaneous localization and mapping for drone navigation in complex environments

This paper introduces a framework for 3-D drone localization and sound source mapping using simultaneous localization and mapping and acoustic techniques. The approach integrates the quaternion method for orientation tracking with a closed-form solution combining the Time Difference of Arrival and Angle of Arrival methods. A drone equipped with three microphone arrays and an inertial measurement unit that captures the drone's linear acceleration and angular velocity simultaneously maps stationary sound sources and estimates its own trajectory. Weighted least squares optimization refines source positions, while particle filtering enhances accuracy in dynamic settings. Simulations show that the method outperforms conventional techniques like dead reckoning, presenting accurate localization and mapping even in complex scenarios. This work highlights the potential of sound-based mapping in challenging scenarios where traditional visual methods fail, such as environments with poor visibility, like underwater or extreme acoustic interference.