Metraipolis › archive › dc_c129efbb5e27

ForceBench: Benchmarking Force Integration for Robotic Manipulation — arXiv:2610.19604

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.19604
License (as archived): CC0-1.0
Archived: 2026-10-15T21:57:54.255Z
SHA-256 of the archived text: ca2cbced38ceb4d5680e6d603a9640081e51cfd51ae4b2046aa8cc8691078a57
Stable URL of this page: https://metraipolis.com/documento/dc_c129efbb5e27

How to cite

ForceBench: Benchmarking Force Integration for Robotic Manipulation — arXiv:2610.19604 — arXiv (en), https://arxiv.org/abs/2610.19604 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c129efbb5e27 (Metraipolis archive dc_c129efbb5e27, sha256 ca2cbced38ceb4d5680e6d603a9640081e51cfd51ae4b2046aa8cc8691078a57)

BibTeX

@misc{dcc129efbb5e27,
  title = {ForceBench: Benchmarking Force Integration for Robotic Manipulation — arXiv:2610.19604},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19604},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c129efbb5e27; sha256 ca2cbced38ceb4d5680e6d603a9640081e51cfd51ae4b2046aa8cc8691078a57},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19604. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Force feedback provides vision-language-action (VLA) policies with physical observations that can guide action selection and execution. However, differences in tasks, demonstrations, and control settings across existing systems hinder reliable comparisons of force integration designs. We introduce ForceBench, a simulation benchmark for evaluating six-axis wrist-force integration under a unified protocol. ForceBench combines eight tasks spanning physical-property inference, resistance-dependent interaction, and contact-rich assembly with 1,600 synchronized demonstrations and task-specific success criteria. We compare one no-force baseline and six force integration designs on a shared $\pi_{0.5}$ backbone. Prompt Fusion achieves the highest observed mean simulation success of 35.31%, compared with 23.25% for the baseline, and improves success rates across all three task categories. Predictive and temporal gains in simulation concentrate on physical-property inference, while both temporal designs underperform the baseline on assembly. Real-world evaluation of eight systems across five tasks supports the main qualitative trends: Prompt Fusion and Dual-Path Fusion achieve 63.00% mean success, compared with 40.00% for the baseline, but only Prompt Fusion improves every task. These findings support ForceBench's practical value for identifying the benefits and limitations of force integration across manipulation demands. The project page is available at 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/2610.19604 — 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 2026-10-15T21:38:44.463Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).