arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.40565cs.ROcs.AIcs.LG
收藏

面向机器人学的Gym-Anything:通过数字智能体扩展物理模拟

Gym-Anything for Robotics: Scaling physical simulations with digital agents

Carnegie Mellon University(卡内基梅隆大学)

Pranjal Aggarwal, Sean Welleck

AI总结:

该研究推出面向机器人学的Gym-Anything框架,利用编码智能体创建机器人模拟,构建含200个RL环境的RoboWorld基准,提出多种前沿模型改进方法,显著提升模型在测试任务中的性能。

摘要:

逼真且快速的机器人模拟可支持大规模训练具身AI模型,使其能迁移至真实世界。然而尽管历经数十年发展,现有模拟仅覆盖狭窄范围的真实世界任务与物理现象。核心瓶颈在于创建更优质的模拟需要大量专家投入,因此难以扩展。为解决该问题,我们推出面向机器人学的Gym-Anything框架,其中编码智能体可创建逼真的机器人模拟。我们的核心见解是模拟创建属于编码智能体任务,因此编码智能体的改进可直接转化为更丰富、更精准的模拟。我们基于美国GDP数据选择模拟以覆盖具有经济价值的物理作业,定义优质模拟与优质任务的标准,并开发对抗性创建-审核循环,依据这些标准对每个模拟与任务进行检查。通过该流程,我们构建了RoboWorld,包含25个模拟,对应25个测试RL环境与175个不同难度的训练RL环境。即便最优前沿模型在所有RoboWorld测试任务中均失败,平均奖励仅为0.06。基于RoboWorld,我们提出前沿模型的多种改进方式,包括跨重复尝试的上下文学习、读取模拟器状态并在上下文中指导学生的特权教师、从模型自身特权经验中学习的自辅助学习,以及生成跨多样场景演示以训练π₀.₅等视觉-语言-动作模型(VLAs)的代码策略。每种方式均显著提升性能。我们发布该流程、RL环境与基准以推动未来研究,代码与数据可访问此https URL。

英文摘要:

Realistic, fast robotics simulations could enable large-scale training of embodied AI models that transfer to the real world. However, despite decades of progress, available simulations cover only a narrow range of real-world tasks and physical phenomena. The core bottleneck is that creating better simulations requires significant expert effort and therefore does not scale. To address this, we introduce Gym-Anything for Robotics, a framework in which coding agents create realistic robotics simulations. Our core insight is that simulation creation is a coding agent task, so improvements in coding agents translate directly into richer and more faithful simulations. We select simulations based on U.S. GDP data to cover economically valuable physical work, define what makes a good simulation and a good task, and develop an adversarial creation-audit loop that checks each simulation and task against these definitions. Using this pipeline, we build RoboWorld, a set of 25 simulations with 25 test and 175 training RL environments of varying difficulty. Even the best frontier models fail on every RoboWorld test task, achieving an average reward of only 0.06. Building on RoboWorld, we introduce several new ways for frontier models to improve, including in-context learning across repeated attempts, a privileged teacher that reads the simulator state and teaches a student in context, self-assisted learning from the model's own privileged experience, and code policies that generate demonstrations across varied scenes to train vision-language-action models (VLAs) such as $π_{0.5}$. Each of these substantially improves performance. We release our pipeline, RL environments, and benchmark to facilitate future research. Code and data are available at https://gym-anything.github.io/robotics

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑