用大模型自动生成机器人行走指令，让四足机器人更智能地适应各种速度。
From LLM-Generated Specifications to Learned Quadruped Locomotion
- 大模型根据自然语言指令生成可执行的逻辑规范
- 在0.3至2.1米/秒全速范围内实现100%成功率和生存率
- 适合对机器人控制自动化有需求的研究者
四足机器人步态策略通常依赖强化学习，而后者严重依赖人工设计的奖励函数。设计奖励函数需大量工程投入，且局部奖励与全局行为之间的关系不明确。通过信号时序逻辑（STL）等形式化规范生成的结构化奖励更具可解释性，但编写STL仍需领域知识。本文研究大语言模型（LLM）能否填补该空白：给定自然语言行走目标和受限语法，GPT-5.5与Qwen 3.6独立生成用于追踪指令、安全性和步态结构的参数化信号时序逻辑（PSTL）模板。通过专家轨迹实例化参数，并保留与专家行为一致的规范。将所得规范转化为平滑的有限历史奖励函数，使用近端策略优化（PPO）在MuJoCo XLA（MJX）中训练四足步态策略。评估了“步态感知”与“步态无感知”两种设置。前者指定行走-小跑、小跑、跳跃等步态区间，后者允许接触模式由任务目标自发形成。相比手工奖励、Text2Reward式代码生成及专家切换基准，步态感知的Qwen 3.6在0.3–2.1米/秒全速范围均达到100%生存率与指令成功率，高精度匹配目标步态；而Text2Reward在≥1.9米/秒时两项指标均为0%。视频演示见https://stl-locomotion.github.io/
原文摘要 · Abstract (English)
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: https://stl-locomotion.github.io/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。