Robust manipulation via denoising diffusion over action sequences.
Diffusion-policy implementation that generates action sequences instead of single actions, dramatically improving robustness on contact-rich and multi-modal tasks. Trained on real-to-sim captured environments.
Hugging Face LeRobot framework — train and deploy with one config.
Open demo →ManipulationCluttered-bin manipulation with F/T wrist sensor on Max.
Open demo →TeleopPredefined teleop dashboard with dual cameras, 3D model, map, and controls.
Open demo →