PhD Position: Distributed Multimodal Learning for Cooperative Acoustic Source Localization

This PhD project focuses on advancing cooperative acoustic source localization and classification in complex environments such as industrial sites or public spaces. The research aims to develop a system where a network of mobile platforms, each equipped with audio-visual sensors, shares observations to reconstruct the 3D position of sound events. The candidate will investigate multimodal learning techniques to ensure the system remains robust against occlusions and partial platform failures. This position is hosted at the Laboratoire Instrumentation Intelligente, Distribuée et Embarquée at Université Paris-Saclay.

• Design and implement algorithms for cooperative 3D acoustic source localization.
• Develop multimodal learning models to integrate audio and visual data from multiple mobile platforms.
• Create communication protocols for sharing directional estimates between neighboring robots.
• Evaluate the robustness of the localization system against occlusions and sensor failures.
• Conduct experiments on mobile platforms to validate the proposed cooperative framework.