Metraipolis › archive › dc_68e00fcb5b70

"Your Robot Was Trained on a Lie": Collision Mesh Poisoning Attacks on Robotic Manipulation — arXiv:2609.18122

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18122
License (as archived)
CC0-1.0
Archived
2026-10-10T05:25:55.179Z
SHA-256 of the archived text
a89e3d92fc21ce2b5fe734c40f0f8b7a6c8558f324e5800d38c6b52c545dd528
Stable URL of this page
https://metraipolis.com/documento/dc_68e00fcb5b70

How to cite

"Your Robot Was Trained on a Lie": Collision Mesh Poisoning Attacks on Robotic Manipulation — arXiv:2609.18122 — arXiv (en), https://arxiv.org/abs/2609.18122 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_68e00fcb5b70 (Metraipolis archive dc_68e00fcb5b70, sha256 a89e3d92fc21ce2b5fe734c40f0f8b7a6c8558f324e5800d38c6b52c545dd528)
BibTeX
@misc{dc68e00fcb5b70,
  title = {"Your Robot Was Trained on a Lie": Collision Mesh Poisoning Attacks on Robotic Manipulation — arXiv:2609.18122},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18122},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_68e00fcb5b70; sha256 a89e3d92fc21ce2b5fe734c40f0f8b7a6c8558f324e5800d38c6b52c545dd528},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18122. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learning-enabled robotic manipulation increasingly relies on robot simulators for policy training and evaluation before real-world deployment. Inside a simulator, a 3D asset contains two separate geometries: a visual mesh used for rendering and a collision mesh used for physical interaction. For computational efficiency, the collision mesh is deliberately a coarse approximation that need not have the same geometry as the visual mesh, a legitimate and pervasive discrepancy we call the Visual--Collision Gap (V--C Gap). We show that the V--C Gap opens a new and practical attack surface, and propose Collision Mesh Poisoning (CMP), the first poisoning attack against robotic manipulation delivered through the 3D asset supply chain. An attacker modifies only the collision mesh of a 3D asset, leaving the visual mesh and all other components unchanged. A policy trained and evaluated with the poisoned asset behaves normally throughout simulation, yet degrades, fails, or creates physical safety risks once deployed in the real world. Since current asset review practices cover malware, copyright, and format compliance, but not visual--collision consistency, poisoned assets can be distributed through legitimate supply chain channels. We evaluate several defenses and our results show that they are insufficient to defend against CMP, highlighting the need for new defenses.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-18 in cs.RO).
Categorie arXiv: Cryptography and Security (cs.CR) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.18122 — 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 2026-10-10T04:47:38.944Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).