Metraipolis › archive › dc_fdbd64e4b455

ContactGuard: Pre-Contact Execution Monitoring with Action-Conditioned Latent World Models — arXiv:2608.13438

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.13438
License (as archived): CC0-1.0
Archived: 2026-10-19T04:18:30.922Z
SHA-256 of the archived text: a0aa2870b5a9f369ce86298a013315d002eaf685b8eac12063d111ee52224f90
Stable URL of this page: https://metraipolis.com/documento/dc_fdbd64e4b455

How to cite

ContactGuard: Pre-Contact Execution Monitoring with Action-Conditioned Latent World Models — arXiv:2608.13438 — arXiv (en), https://arxiv.org/abs/2608.13438 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fdbd64e4b455 (Metraipolis archive dc_fdbd64e4b455, sha256 a0aa2870b5a9f369ce86298a013315d002eaf685b8eac12063d111ee52224f90)

BibTeX

@misc{dcfdbd64e4b455,
  title = {ContactGuard: Pre-Contact Execution Monitoring with Action-Conditioned Latent World Models — arXiv:2608.13438},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13438},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fdbd64e4b455; sha256 a0aa2870b5a9f369ce86298a013315d002eaf685b8eac12063d111ee52224f90},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13438. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Contact-rich manipulation failures are often detected only after the robot has committed to contact. This is especially limiting in wrist-camera setups: close gripper--object views help observe contact, but a poor approach may already push, miss, slip, or disturb the object before conventional detectors react. We introduce \emph{ContactGuard}, a pre-contact execution monitor for chunked visuomotor policies. Given the policy's planned action chunk, ContactGuard predicts its short-horizon consequence in latent visual space and aborts if the predicted future latent indicates likely failure. Its latent world model is trained from unlabelled robot trajectories to predict compact multi-view visual embeddings under planned actions, avoiding pixel-level video prediction. A lightweight failure probe is then trained from a small labelled set of pre-contact clips. At deployment, ContactGuard anchors prediction before an imminent contact event, rolls the model forward under the policy's own actions, and verifies the predicted post-contact latent. Across real-world contact-rich manipulation tasks, ContactGuard predicts failure more accurately than direct and corrupted-action ablations, and transfers to live robot as a pre-contact abort signal without modifying the underlying policy.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.AI).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.13438 — 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 2026-10-19T04:13:06.770Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).