Metraipolis › archive › dc_bc6afa774a65
ELEVATE: Agentic Controller Acquisition from Task Failure for Quadrupedal Manipulation — arXiv:2610.38388
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.38388
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-01T02:44:13.535Z
- SHA-256 of the archived text
7d5bdc46b1c37d3f3690a030611a0d2d0353260178f8d8adf00b893198ffdc76- Stable URL of this page
https://metraipolis.com/documento/dc_bc6afa774a65
How to cite
ELEVATE: Agentic Controller Acquisition from Task Failure for Quadrupedal Manipulation — arXiv:2610.38388 — arXiv (en), https://arxiv.org/abs/2610.38388 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bc6afa774a65 (Metraipolis archive dc_bc6afa774a65, sha256 7d5bdc46b1c37d3f3690a030611a0d2d0353260178f8d8adf00b893198ffdc76)
BibTeX
@misc{dcbc6afa774a65,
title = {ELEVATE: Agentic Controller Acquisition from Task Failure for Quadrupedal Manipulation — arXiv:2610.38388},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.38388},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bc6afa774a65; sha256 7d5bdc46b1c37d3f3690a030611a0d2d0353260178f8d8adf00b893198ffdc76},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38388. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Quadrupedal manipulators can extend their workspace through whole-body interaction, but task-planning agents' capabilities remain limited by their available controllers. We present ELEVATE, a framework that uses persistent failures during task-program refinement to identify and acquire missing whole-body control capabilities. When these failures suggest a missing physical capability, a coding agent formulates and iteratively refines a reinforcement-learning problem, specifying behavior-specific rewards, curricula, and termination conditions. Validated controllers are added to the capability library and composed with existing controllers to retry the task. Starting from a single pretrained loco-manipulation controller, ELEVATE autonomously acquires controllers for wall-supported elevation, supported manipulation, and wall descent across a sequence of seven simulation tasks. These controllers are reused in subsequent reaching and object-manipulation tasks. On the five tasks beyond the initial controller's workspace, ELEVATE achieves mean success rates of 98.4%-100% across five independent runs, while coding-agent baselines with a fixed controller library fail to complete these tasks. The synthesized task programs, together with the acquired controllers, also transfer to a physical quadrupedal manipulator without real-robot training or fine-tuning, achieving 18/20 successes in rack retrieval and 16/20 in floor-to-rack placement. Videos and demonstrations are 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/2610.38388 — 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 2026-11-01T02:16:58.627Z dalla pagina https://arxiv.org/list/cs.RO/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).