SOURCE-LINKED INTELLIGENCE

Uranus: Building the Next-Generation Simulation Infrastructure for Embodied AI

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Scalable simulation is essential for robot data generation, policy training, evaluation, and safe iteration, yet real-world interaction is costly and conventional simulators require labor-intensive construction. We present Uranus, a data-driven robot simulator built around a joint-trajectory-conditioned autoregressive diffusion model. Uranus offers three key capabilities: (1) streaming, open-ended rollout, which receives future joint-position trajectories online and autoregressively generates one latent frame per step, corresponding to four RGB frames, without a fixed horizon; (2) low-latency

Read original source ↗ Open in workspace

recordType

paper

region

Global