About this document
Reactive Diffusion Policy - Slow-Fast Visual-Tactile Policy Learning For Contact-Rich Manipulation by Zijian Wang is a document available to read on EtoBox.
The document introduces TactAR, a low-cost teleoperation system that provides real-time tactile feedback through Augmented Reality (AR), and the Reactive Diffusion Policy (RDP), a slow-fast visual-tactile imitation learning algorithm designed for complex contact-rich manipulation tasks. RDP utilizes a two-level hierarchy with a slow latent diffusion policy for high-level action prediction and a fast asymmetric tokenizer for closed-loop control based on tactile feedback. Experiments demonstrate that RDP sign
- Author
- Zijian Wang
- Language
- EN