Metraipolis › archive › dc_428a5b00166f

Social-WM: Safety-Aware Latent World Models for Robot Social Navigation — arXiv:2609.40177

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.40177
License (as archived)
CC0-1.0
Archived
2026-10-07T15:12:21.194Z
SHA-256 of the archived text
88c3ddafaf3c5e648a9a8777b901127faefa6ea4c8bc4a27b67eff189040a86f
Stable URL of this page
https://metraipolis.com/documento/dc_428a5b00166f

How to cite

Social-WM: Safety-Aware Latent World Models for Robot Social Navigation — arXiv:2609.40177 — arXiv (en), https://arxiv.org/abs/2609.40177 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_428a5b00166f (Metraipolis archive dc_428a5b00166f, sha256 88c3ddafaf3c5e648a9a8777b901127faefa6ea4c8bc4a27b67eff189040a86f)
BibTeX
@misc{dc428a5b00166f,
  title = {Social-WM: Safety-Aware Latent World Models for Robot Social Navigation — arXiv:2609.40177},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.40177},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_428a5b00166f; sha256 88c3ddafaf3c5e648a9a8777b901127faefa6ea4c8bc4a27b67eff189040a86f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40177. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Safe social navigation requires a robot to anticipate not only the future consequences of its actions, but also whether a nominal action can actually be executed under surrounding physical and social constraints. We present Social-WM, an efficient latent world-model planning framework trained from egocentric RGB video sequences. Our key observation is that social-navigation experience contains a systematic discrepancy between the nominal action and the realizable action: a nominal forward action may be fully executed in free space, but needs to be constrained when heading towards a pedestrian or obstacle. Social-WM learns these safety-relevant consequences directly through action-conditioned future prediction, where the target is the actual observed future following each command. We further introduce a realizable inverse-dynamics objective that associates observed latent transitions with the action actually realized rather than the nominal one. At deployment, candidate actions are imagined through the latent world model, and the inverse dynamics model estimates their realizability; nominal--realizable discrepancy then provides a safety signal before execution. The learned dynamics and realizability model remain goal-independent and support both position- and image-goal navigation. On Social-HM3D, Social-WM achieves 63.77% success while reducing human collisions to 21.67%, and maintains strong performance under zero-shot transfer to Social-MP3D, without explicit pedestrian tracking, privileged human state, or online reinforcement learning.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-01 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.40177 — 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 2026-10-07T14:44:47.495Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).