Metraipolis › archive › dc_e980b658d000

SimVLA: Zero-Shot Sim-to-Real VLA Learning for Mobile Manipulation — arXiv:2610.11248

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.11248
License (as archived)
CC0-1.0
Archived
2026-10-09T06:33:15.925Z
SHA-256 of the archived text
834775944539e34a1ecd2a71abfcd36aa305fa28c1a226e51775da8b76d1561c
Stable URL of this page
https://metraipolis.com/documento/dc_e980b658d000

How to cite

SimVLA: Zero-Shot Sim-to-Real VLA Learning for Mobile Manipulation — arXiv:2610.11248 — arXiv (en), https://arxiv.org/abs/2610.11248 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e980b658d000 (Metraipolis archive dc_e980b658d000, sha256 834775944539e34a1ecd2a71abfcd36aa305fa28c1a226e51775da8b76d1561c)
BibTeX
@misc{dce980b658d000,
  title = {SimVLA: Zero-Shot Sim-to-Real VLA Learning for Mobile Manipulation — arXiv:2610.11248},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.11248},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e980b658d000; sha256 834775944539e34a1ecd2a71abfcd36aa305fa28c1a226e51775da8b76d1561c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.11248. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large-scale, diverse datasets have driven the success of LLMs and VLMs. But VLAs for robotics remain limited by the cost and complexity of real-world data collection. While simulation offers a scalable alternative, its potential for sim-to-real VLA learning in mobile manipulation remains largely underexplored. We introduce SimVLA, an end-to-end framework that trains VLAs entirely on synthetic simulation data without teleoperation for mobile manipulation. SimVLA is first pre-trained on two complementary simulation-derived datasets: SimAction, a large-scale robot action dataset spanning 35 diverse mobile manipulation tasks, generated by composing atomic skills, and SimVQA, which leverages privileged simulator state to provide spatial, geometric, and subtask-level visual-language supervision. We further post-train SimVLA on a mixture of SimAction and SimDeploy, a dataset collected from policy rollouts across diverse simulated environments. We evaluate SimVLA on tasks including restocking, pouring, and cleaning, and show zero-shot transfer to real-world mobile manipulation, including real home environments. SimVLA outperforms policies trained on 50 in-domain real-world demonstrations, suggesting that simulation can enable scalable sim-to-real mobile manipulation. We further demonstrate the value of multiple complementary forms of supervision for effectively leveraging simulation in VLA training.

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/2610.11248 — 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 2026-10-09T05:55:22.013Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).