Event
DEX-PHD: A Portable Handheld Dexterous Manipulation Data Collection Device
1 reports · 1 sources
Jooyoung Hong and other authors released a paper proposing a portable handheld device, DEX-PHD, for collecting multimodal data on dexterous manipulation. The device integrates a dexterous robotic hand, per-finger continuous control, and multimodal recording, and can collect proprioceptive information, fingertip contact data, and wrist camera video; the operator directly controls the robotic hand through joint-level mapping. Its cable-free design supports data collection in a variety of environments. The authors demonstrated the device on contact-intensive tasks requiring multi-finger coordination, and executed the collected demonstrations on a physical humanoid robot via open-loop replay; the results show that open-loop replay preserved the main temporal patterns of finger motion and task-relevant contacts.
Report timeline
- DEX-PHD: A Portable Handheld Device for Multimodal Data Collection for Dexterous Manipulation
The authors propose DEX-PHD, a portable handheld device that integrates a dexterous robot hand, continuous per-finger control, and multimodal logging, and can collect proprioceptive information, fingertip contact data, and wrist-camera video, with the operator directly controlling the robot hand through joint-level mapping. Its untethered design supports data collection in a variety of environments. The authors demonstrate it on contact-rich tasks that require multi-finger coordination, and execute the collected demonstrations on a physical humanoid robot through open-loop replay; the results show that open-loop replay preserves the main temporal patterns of finger motion and task-relevant contact.
arXiv cs.RO ·