Metraipolis › archive › dc_3979edfad77e
From LLM-Generated Specifications to Learned Quadruped Locomotion — arXiv:2609.07111
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.07111
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T15:31:02.854Z
- SHA-256 of the archived text
4f1497f61756ec2cfada9cde0040d512909f3ad008d4af5ec47c11676c2c06d4- Stable URL of this page
https://metraipolis.com/documento/dc_3979edfad77e
How to cite
From LLM-Generated Specifications to Learned Quadruped Locomotion — arXiv:2609.07111 — arXiv (en), https://arxiv.org/abs/2609.07111 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3979edfad77e (Metraipolis archive dc_3979edfad77e, sha256 4f1497f61756ec2cfada9cde0040d512909f3ad008d4af5ec47c11676c2c06d4)
BibTeX
@misc{dc3979edfad77e,
title = {From LLM-Generated Specifications to Learned Quadruped Locomotion — arXiv:2609.07111},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.07111},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3979edfad77e; sha256 4f1497f61756ec2cfada9cde0040d512909f3ad008d4af5ec47c11676c2c06d4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07111. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Quadruped robot locomotion policies are often trained using reinforcement learning, which in turn relies heavily on hand-crafted reward functions. Designing reward functions requires substantial manual engineering, and it is often unclear which local rewards will induce the desired global behavior. Shaped rewards from formal specifications in languages like Signal Temporal Logic (STL) can make rewards more interpretable, but writing STL specifications itself still requires domain expertise. We study whether large language models (LLMs) can fill this gap by generating Parametric Signal Temporal Logic (PSTL) specifications that are subsequently used for policy learning. Given a natural language locomotion objective and a constrained specification grammar, GPT-5.5 and Qwen 3.6 independently propose STL templates for command tracking, safety, and gait structure. We instantiate the parameters of the generated PSTL templates using expert trajectories and retain only specifications that are consistent with demonstrated expert behavior. The resulting specifications are then transformed into smooth, finite-history reward functions and used to train a quadruped locomotion policy with Proximal Policy Optimization (PPO) in MuJoCo XLA (MJX). We evaluate both \emph{gait-aware} and \emph{gait-agnostic} settings. The former specifies walking-trot, trot, and bound regimes, while the latter allows contact patterns to emerge from the task objective. We compare against hand-engineered rewards, Text2Reward-style LLM-generated reward code, and an expert-switching oracle. Gait-aware Qwen 3.6 specifications achieved 100\% survival and command success across all tested speeds (0.3--2.1 m/s) and matched the target gait at high speeds, whereas Text2Reward achieved 0\% for both metrics at $\geq 1.9$ m/s. Videos: this https URL
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.RO).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.07111 — 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 2026-10-16T15:11:47.568Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).