DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity
DexFLEX is a contact-aware foundation controller that transforms fingertip motion drafts into contact-consistent joint commands.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
policy-learning success
3.5
policy-learning success
1.56
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
for responsiveness, then decodes the selected motion into executable joint-space targets. Across degraded-command simulation, real-world shared control, and visuomotor policy learning, the same trained controller improves robustness without retraining, increasing real-world teleoperation success from 29.2% to 78.3% and reaching 46.7% policy-learning success, 3.5 times direct joint-action prediction, and 1.56 times pr
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org