LATIDIA · Robótica

Habitat-GS: un simulador de navegación de alta fidelidad con splatting gaussiano dinámico

arXiv: 2604.12626v2Tipo de anuncio: reemplazar Resumen: El entrenamiento de los agentes de IA incorporados depende críticamente de la fidelidad visual de los entornos de simulación y la capacidad de modelar humanos dinámicos. Simuladores actuales predominantes

La noticia

arXiv:2604.12626v2 Announce Type: replace Abstract: Training embodied AI agents depends critically on the visual fidelity of simulation environments and the ability to model dynamic humans. Current simulators predominantly rely on mesh-based rasterization, for which photorealistic assets are costly to author at scale, and their support for dynamic human avatars is largely constrained to mesh representations, hindering agent generalization to human-populated real-world scenarios. We present Habitat-GS, a navigation-centric embodied AI simulator extended from Habitat-Sim that integrates 3D Gaussian Splatting scene rendering and drivable gaussian avatars while maintaining full compatibility with the Habitat ecosystem. Our system implements a 3DGS renderer for real-time photorealistic rendering and supports scalable 3DGS asset import from diverse