Robots Learn to See, Feel and Hear at Once as Multim,odal Perception Advances

A collection of five studies explores advancements in multimodal perception for robotics, integrating vision, touch, and sound to create a coherent understanding of the environment. The research addresses challenges in sensor fusion, including temporal and spatial discrepancies, and introduces novel frameworks for tactile sensing, cross-modal alignment, and structured reasoning. The work emphasizes the need for robust, interpretable representations in safety-critical applications.