Metraipolis › archive › dc_049d3fc30db4

CAP: Continuously Adaptive Perception-Blind Humanoid Locomotion via Learned Denoising — arXiv:2609.11553

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11553
License (as archived)
CC0-1.0
Archived
2026-10-16T06:36:02.372Z
SHA-256 of the archived text
21fafb8a4a899634ee6088e50c777f230d3ade75fcae3021599bbdf7410de981
Stable URL of this page
https://metraipolis.com/documento/dc_049d3fc30db4

How to cite

CAP: Continuously Adaptive Perception-Blind Humanoid Locomotion via Learned Denoising — arXiv:2609.11553 — arXiv (en), https://arxiv.org/abs/2609.11553 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_049d3fc30db4 (Metraipolis archive dc_049d3fc30db4, sha256 21fafb8a4a899634ee6088e50c777f230d3ade75fcae3021599bbdf7410de981)
BibTeX
@misc{dc049d3fc30db4,
  title = {CAP: Continuously Adaptive Perception-Blind Humanoid Locomotion via Learned Denoising — arXiv:2609.11553},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11553},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_049d3fc30db4; sha256 21fafb8a4a899634ee6088e50c777f230d3ade75fcae3021599bbdf7410de981},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11553. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Humanoid locomotion across complex terrain demands forward-looking exteroception to anticipate obstacles, yet this signal is unreliable in real-world deployment, failing partially and intermittently. Existing perceptive policies often assume that depth observations remain clean and in-distribution, while recent attempts to unify perceptive and blind control typically route or switch between separate sub-policies, leaving recoverable information in partially corrupted depth unexploited. We instead propose CAP, a single-stage humanoid locomotion policy that recovers this signal with a perceptive world-model encoder trained as a learned denoiser to reconstruct clean depth from a corrupted input, together with a co-active proprioceptive variational encoder that supplies depth-free body-state information. A coupled training recipe pairs a depth-noise curriculum on the world-model input with world-model feature dropout on the policy-facing latent, exposing the policy to failures across the entire perception-quality spectrum. In simulation, CAP matches or improves upon perceptive baselines when depth remains informative, and degrades more smoothly than a binary-switching baseline as perception worsens. On the Unitree G1, controlled trials and indoor-outdoor deployments demonstrate perception-robust locomotion under intermittent occlusion, real-sensor corruption, and outdoor depth artifacts.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-11 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.11553 — 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 2026-10-16T06:10:18.232Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).