EmbodiedLab: Benchmarking Embodied Agents That Act to Know
作者: Hongcheng Gao, Zelin Zheng, Alex Zhou, Jay Zhu, Shijia Ge, Yazhe Wang, Xiping Li, Jianshu Zeng, Di Wu, Xuan Shangguan, Yexing Wan, Lingyu He, Yongli Wang, Xiaohan Yu, Zhiqi Jia, Yue Liu, Bryan Hooi, Yinpeng Dong, Wentao Zhang, Sihang Wu, Xiao He
分类: cs.RO
发布日期: 2026-10-26
备注: Technical report. 19 pages, 10 figures, 9 tables. Project page: https://hexafuture.ai/benchmarks/embodiedlab/
🔗 代码/项目: GITHUB
💡 一句话要点
提出EmbodiedLab基准以解决机器人自主探索问题
🎯 匹配领域: 支柱一：机器人控制 (Robot Control)
关键词: 机器人探索 自主决策 科学发现 物理实验 操作技能 知识获取 动态参数
📋 核心要点
- 现有的操作基准未能有效评估机器人在探索未知环境中的自主决策能力，限制了其在科学发现中的应用。
- 本文提出EmbodiedLab基准，要求机器人在物理实验中通过自主操作获取知识，强调知识获取过程的重要性。
- 在30个实验中，评估了六种前沿模型的表现，结果显示操作技能对弱代理的提升效果显著，且大多数失败源于与设备的互动。
📝 摘要（中文）
Embodied agents期望通过自主决策与环境互动来实现现实世界目标。现有的操作基准仅关注状态改变，而科学发现基准则用符号动作替代物理实验。本文提出EmbodiedLab基准，要求机器人通过操作来获取知识。该基准涵盖30个实验，涉及五个学科和14个设备平台，机器人需选择操作程序并在物理模拟实验室中进行实验。评估标准包括机器人所获得的知识和任务所需的物理状态。实验结果显示，尽管最强代理的成功率不足50%，但操作技能对较弱代理的帮助显著。
🔬 方法详解
问题定义：本文旨在解决现有基准无法有效评估机器人自主探索与知识获取能力的问题。现有方法往往只关注状态改变或用符号替代物理实验，缺乏对知识获取过程的真实评估。
核心思路：提出EmbodiedLab基准，要求机器人在物理实验中自主选择操作程序，通过与环境的互动来获取知识。这种设计强调了探索过程的重要性，而不仅仅是最终结果。
技术框架：EmbodiedLab基准包含多个模块，包括任务设定、操作选择、实验执行和知识评估。机器人在每个实验中接收目标，选择操作程序，进行实验并记录结果。
关键创新：最重要的创新在于将知识获取与物理实验结合，机器人不仅要改变环境状态，还需观察反馈以获得知识。这与传统方法的本质区别在于强调探索过程而非仅仅关注结果。
关键设计：实验中采用了动态参数重采样，确保每次实验的结果不可预知。评估标准包括机器人所获得的知识和任务所需的物理状态，确保每个实验都是可解的。
🖼️ 关键图片
📊 实验亮点
实验结果显示，尽管最强代理的成功率不足50%，但操作技能显著提升了较弱代理的表现。大多数失败发生在与设备的互动中，表明操作技能在自主探索中的关键作用。
🎯 应用场景
该研究的潜在应用领域包括机器人科学研究、教育和工业自动化等。通过提升机器人在未知环境中的自主探索能力，能够促进科学发现和技术创新，未来可能在智能制造和自主系统中发挥重要作用。
📄 摘要（原文）
Embodied agents are expected to accomplish real-world goals by deciding for themselves how and when to interact with their surroundings. For many such goals, changing the physical state of the world is the means rather than the end: the agent changes the state of a system through interaction, observes the feedback, and explores until it either achieves the goal itself (e.g., discovering physical laws) or learns what the goal requires; it acts to know. Manipulation benchmarks stop at acting to change, with the target state given; scientific discovery benchmarks evaluate knowing but replace physical experimentation with symbolic actions or numerical queries. We introduce EmbodiedLab, a benchmark in which a robot must act to know. Quantitative science experiments make this setting controllable: the parameters of the system are resampled for every instance, so the result cannot be known in advance, yet it can be scored exactly. Across 30 experiments in five disciplines on 14 apparatus platforms, the agent receives an objective and the apparatus, chooses its own procedure, and operates a physics-based simulated laboratory, reading the instruments through its cameras. Evaluation checks what the agent came to know, grounded in readings recorded during the episode, as well as the physical state the task requires; every experiment is verified to be solvable. We evaluate six frontier models in their developers' coding agents with native observation and control, with added perception and motion planning, and with manipulation skills distilled from the strongest agents. With native control, even the strongest agent succeeds less than half the time; for every agent, most failures arise in interacting with the apparatus, and manipulation skills help weaker agents more than perception and planning. The benchmark and its evaluation code are available at https://github.com/HexaFuture/EmbodiedLab.