ManipulationHardware: MaxRoadmap scaffold
Warehouse pick-and-stow
Cluttered-bin manipulation with F/T wrist sensor on Max.
Lab runtime / ManipulationOK / VIS-01
Warehouse pick-and-stow runs against the same observable edge stack used for testing and deployment.
This entry is a roadmap scaffold, not a finished demo. Its description records intended scope; the linked repository is the source of truth for current code.
About this demo
Industrial demo of cluttered-bin picking with closed-loop force-torque feedback. Uses the optional ATI Mini45 F/T payload on the Kinematics Max box. Trained from real-to-sim capture of a customer warehouse cell.
Highlights
- →F/T wrist payload (optional)
- →Cluttered-bin grasping
- →Real-to-sim trained
- →Vision-fallback re-grasp
Supported robots
UR10eFranka Panda
Related demos
View all →Manipulation
Diffusion Policy
Robust manipulation via denoising diffusion over action sequences.
Open demo →ManipulationLeRobot ACT policy
Hugging Face LeRobot framework — train and deploy with one config.
Open demo →TeleopCockPit
Predefined teleop dashboard with dual cameras, 3D model, map, and controls.
Open demo →