MiTaS: Multi-Resolution Tactile Fusion Hits 80% Success on Contact-Rich Robot Tasks
GeorgiaChal · x · 2026-10-01
MiTaS (Multi-Resolution Tactile Sensing) from TU Darmstadt, accepted at CoRL 2026 and a finalist for two IROS 2026 workshop awards, fuses an RGB camera, a GelSight Mini vision-based sensor, and a high-frequency Evetac event-based sensor.
Method: modality-specific convolutional stems plus transformer-based fusion condition a flow-matching policy for downstream manipulation.
Results: averaging 80% success across five real-world contact-rich tasks, beating vision-only and standard visuo-tactile baselines — multi-resolution tactile features provide real gains for imitation learning.
More from Embodied
- Unitree T-series humanoid shows off impressive kip-up move — jonstephens85 · 2026-10-01
- Dyna-2.1 Breakdown: How the Team Built the First Physical Agent With 1-Hour Whole-Body Autonomy — chris_j_paxton · 2026-10-01
- Why RL for Low-Level Robot Control: Dyna-2.1 Author's Three Key Reasons — chris_j_paxton · 2026-10-01
- InSpatio-World 1.5 turns images and videos into real-time 4D worlds, tops WorldScore — Scobleizer · 2026-10-01
- IROS Debate: Generalist Robots Nail Coarse Placement 19/20 but Fail Precision Insertion 2/20 — GeorgiaChal · 2026-10-01
- NVIDIA Researcher Explains Cosmos: A World Foundation Model Is a Foundation Model of World Models — liu_mingyu · 2026-10-01