Metraipolis › archive › dc_052d32d55ed9

$TF^2$-Steering: Action Refinement with Tactile Forecasts and Feedback via Velocity-Field Steering — arXiv:2611.01483

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.01483
License (as archived): CC0-1.0
Archived: 2026-11-03T18:57:09.293Z
SHA-256 of the archived text: 113e7864eab46efdeb00c6139fb600714aa153d0f45da41b9288c1692d87f17a
Stable URL of this page: https://metraipolis.com/documento/dc_052d32d55ed9

How to cite

$TF^2$-Steering: Action Refinement with Tactile Forecasts and Feedback via Velocity-Field Steering — arXiv:2611.01483 — arXiv (en), https://arxiv.org/abs/2611.01483 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_052d32d55ed9 (Metraipolis archive dc_052d32d55ed9, sha256 113e7864eab46efdeb00c6139fb600714aa153d0f45da41b9288c1692d87f17a)

BibTeX

@misc{dc052d32d55ed9,
  title = {$TF^2$-Steering: Action Refinement with Tactile Forecasts and Feedback via Velocity-Field Steering — arXiv:2611.01483},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.01483},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_052d32d55ed9; sha256 113e7864eab46efdeb00c6139fb600714aa153d0f45da41b9288c1692d87f17a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01483. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Tactile sensing plays a crucial role in everyday human interaction. Consider inserting a plug into a tight socket, vision alone provides limited guidance during the final insertion stage, and we instead rely on both anticipated contact and realized tactile feedback to ajust the insertion motion. The same capability is equally important for robots, particularly in dexterous manipulation tasks. Yet many contemporary robot policies remain vision-dominant. This visual bias, compounded by limited tactile data and sparse contact events, can cause directly incorporated tactile signals to be ignored or even compromise the policy's pretrained capabilities, yielding degrading performance as observed in our experiments. We present $TF^2$-Steering, which refines actions via learned residual steering of the flow-matching action expert's velocity field, using a tactile mixture-of-experts architecture to integrate both tactile forecasts and real-time tactile feedback. Together with specific designed three stages training recipe, $TF^2$-Steering enables tactile signals to correct and, when necessary, redirect the flow-matching action-generation process, while building on the strong action-generation capacity of the vision-based foundation policy. Extensive experiments across six simulated and six real-world manipulation tasks demonstrate the effectiveness of $TF^2$-Steering. It outperforms the strongest evaluated baselines by 10.7 percentage points in average simulation success rate and 9.1 percentage points in average real-world progress success rate. Project page: this https URL .

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.RO, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2611.01483 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T18:17:07.731Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).