X Square Robot's TwinDEX collects dexterous data 5.3x faster than teleoperation
heyronir · x · 2026-09-02
The author argues that what limits dexterous robot learning may be the data pipeline, not the model, highlighting X Square Robot's TwinDEX: a human demonstrates dexterous tasks through a wearable manipulation interface while its twinned robot hand learns directly from those demos, keeping the robot out of the loop.
Key numbers: learning efficiency comparable to on-robot teleoperation data, collection throughput 5.3x on-robot teleoperation, with zero on-robot teleop data used in training.
The deeper principle: design the wearable interface and robot hand as a matched system from the start, and you may not need to solve hardware mismatch entirely in software. A direction worth watching for the economics of dexterous data collection.
More from Embodied
- SimInfinity: open-source robotic simulation benchmark with 10K real-world tasks — joemeno · 2026-10-18
- Robi: a German/English voice companion demo built on Gemma, Tyger and Humaneness Voice TTS — realmrfakename · 2026-10-18
- Munich's Plim unveils wheeled home-care humanoid controllable from 9,000 km away — CyberRobooo · 2026-10-18
- Musk says Optimus and Cybercab will make up the vast majority of Tesla's value — XFreeze · 2026-10-18
- SAE weight editing steers robot behavior at inference time, IIIT Hyderabad paper lands IJCAI Safe Physical AI workshop — ponguru · 2026-10-18
- NTU-led survey reviews nearly 200 papers on affordance learning for embodied AI — jiqizhixin · 2026-10-18