One touch encoder generalizes across sensors, cutting contact error to 0.16mm
kimmonismus · x · 2026-11-11
Part 4 of the thread: a single LingBot-Touch encoder reads three very different sensor types — magnetic, visuotactile and piezoresistive. On XHand1, whose sensor model wasn't seen in pretraining, the frozen encoder beat a sensor-specific adapter on all five probe tasks, cutting contact-center error from 5.89mm to 0.16mm.
More from Embodied
- CoRL best-paper nominee Play2Perfect ported to new robots in IsaacLab and MJLab — leto__jean · 2026-11-11
- Walden Robotics lays out its embodied AI playbook: full-stack, sim, teleop, human data — chris_j_paxton · 2026-11-11
- 5,000 hours of paid touch data: robotics' scarcest, unscrapeable training input — kimmonismus · 2026-11-11
- Chris Paxton rides a Zoox robotaxi for the first time since his internship 10 years ago — chris_j_paxton · 2026-11-11
- Robotics VC bull case: Physical AI funding still below pre-ChatGPT levels, winners will win — Rewkang · 2026-11-11
- Binary-reward RL hits 94% success on hard robotics task, recipe looks clear — kastnerkyle · 2026-11-11