Metraipolis › archive › dc_8716d585ceea
FRAMES: Failure Recovery And Monitoring of Embodied Skills for Humanoid Loco-Manipulation — arXiv:2609.22538
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.22538
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T22:21:25.343Z
- SHA-256 of the archived text
9f1325f610a753e6f29930a71921226d23e571ae0661c817fd21a7826b1f0b71- Stable URL of this page
https://metraipolis.com/documento/dc_8716d585ceea
How to cite
FRAMES: Failure Recovery And Monitoring of Embodied Skills for Humanoid Loco-Manipulation — arXiv:2609.22538 — arXiv (en), https://arxiv.org/abs/2609.22538 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8716d585ceea (Metraipolis archive dc_8716d585ceea, sha256 9f1325f610a753e6f29930a71921226d23e571ae0661c817fd21a7826b1f0b71)
BibTeX
@misc{dc8716d585ceea,
title = {FRAMES: Failure Recovery And Monitoring of Embodied Skills for Humanoid Loco-Manipulation — arXiv:2609.22538},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.22538},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8716d585ceea; sha256 9f1325f610a753e6f29930a71921226d23e571ae0661c817fd21a7826b1f0b71},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22538. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language model (LLM) planners can decompose natural-language instructions and select reusable robot skills, but choosing the correct skill does not guarantee successful physical execution. This gap is especially important in humanoid loco-manipulation, where errors during approach, grasping, transport, or placement can invalidate the remainder of a long-horizon plan. We present FRAMES, a failure-aware supervisory framework for the Unitree G1 humanoid that operates above the CEER whole-body controller. A Planner Agent selects subtasks through parameterized mid-level skills, while a vision-language-model-based Monitor Agent evaluates each skill using temporal multi-view observations and structured robot and contact evidence. Detected failures stop the active skill and provide grounded feedback to a Recovery Agent. The framework further includes a Memory Module for reusing prior skill experience, and geometric grounding via depth and segmentation. We independently evaluate the monitoring module of the framework in MuJoCo using 100 trials comprising 50 failed and 50 successful executions across five tasks. The monitor detects 48 of 50 failures, correctly accepts 46 of 50 successful executions, and achieves 94.0% overall accuracy. These results provide initial evidence for the monitoring component, while end-to-end evaluation of the complete recovery loop remains ongoing. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.RO). Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.22538 — 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 2026-10-09T21:59:59.433Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).