Metraipolis › archive › dc_54ca1ce2fa52
Robo-Harness K1: Harnessing Robot-Use Agents via Perception Augmentation — arXiv:2609.29389
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.29389
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T07:16:54.499Z
- SHA-256 of the archived text
fb0a0d297a31844dfa7a3d04e0987e5ed0e5d4858506eae9f0d804501072375d- Stable URL of this page
https://metraipolis.com/documento/dc_54ca1ce2fa52
How to cite
Robo-Harness K1: Harnessing Robot-Use Agents via Perception Augmentation — arXiv:2609.29389 — arXiv (en), https://arxiv.org/abs/2609.29389 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_54ca1ce2fa52 (Metraipolis archive dc_54ca1ce2fa52, sha256 fb0a0d297a31844dfa7a3d04e0987e5ed0e5d4858506eae9f0d804501072375d)
BibTeX
@misc{dc54ca1ce2fa52,
title = {Robo-Harness K1: Harnessing Robot-Use Agents via Perception Augmentation — arXiv:2609.29389},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.29389},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_54ca1ce2fa52; sha256 fb0a0d297a31844dfa7a3d04e0987e5ed0e5d4858506eae9f0d804501072375d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29389. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Foundation vision-language models (VLMs) understand objects, instructions, and spatial relations, yet translating this capability into robotic manipulation remains difficult. Vision-language-action (VLA) models require extensive demonstrations and may compromise pretrained understanding, while direct RGB-only VLM control is costly and strongly dependent on model capability. We introduce Robo-Harness K1, a robot-use agent (RUA) framework that exposes perception as tools. The agent queries calibrated depth, persistent visual anchors, spatial measurements, and grasp hypotheses, then selects generic motions from the returned evidence. This interface makes 3D geometry accessible without changing the VLM architecture or training a depth encoder. On matched LIBERO-PRO tasks, Gemini 3.7 Flash with K1 reaches 77.8% accuracy, surpassing GPT-6 Astra's 61.1% with an RGB-only harness; K1 further improves Astra to 88.9%. Without target fine-tuning, Gemini with K1 transfers to three RoboSuite arms and dual-arm RoboTwin tasks. On RoboTwin, it achieves 32.0% on Easy and 28.0% on Hard, showing resilience to visual and environmental perturbations. K1 also produces tool-call traces aligned with next-token training. A Qwen3.5-9B student trained on only 107 teacher episodes reaches 44.2% accuracy on new initial states versus 30.2% for OpenVLA, and 13.9% on held-out task conditions versus 0.0% for OpenVLA. These results suggest that perception-augmented RUAs offer a promising route to sample-efficient, generalizable robotic policies that leverage VLM capabilities through an accessible tool interface. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-25 in cs.RO). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.29389 — 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 2026-10-08T06:38:53.805Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).