arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.11931cs.LGcs.NEcs.RO
收藏

MetaMorph:使用Transformers学习通用控制器

MetaMorph: Learning Universal Controllers with Transformers

Stanford University(斯坦福大学);NVIDIA Corporation(英伟达公司);Stanford Institute for Human-Centered Artificial Intelligence(斯坦福以人为本人工智能研究院)

Agrim Gupta, Linxi Fan, Surya Ganguli, Li Fei-Fei

摘要:

通过利用Transformers进行大规模预训练,然后进行特定任务的微调,视觉、自然语言和音频等多个领域正在取得巨大进展。相比之下,在机器人技术中,我们主要是为单一任务训练单个机器人。然而,模块化机器人系统现在允许通用构建块到任务优化形态的灵活组合。然而,考虑到可能的机器人形态的数量呈指数级增长,为每个新设计训练控制器是不切实际的。在这项工作中,我们提出了MetaMorph,一个基于Transformer的方法来学习一个通用的控制器在一个模块化的机器人设计空间。MetaMorph基于这样一种见解,即机器人形态只是我们可以调节Transformer输出的另一种形态。通过大量的实验,我们证明了各种机器人形态的大规模预训练的结果与组合泛化能力的政策,包括zero shot泛化到看不见的机器人形态。我们进一步证明,我们的预训练策略可以用于样本高效转移到全新的机器人形态和任务。

英文摘要:

Multiple domains like vision, natural language, and audio are witnessing tremendous progress by leveraging Transformers for large scale pre-training followed by task specific fine tuning. In contrast, in robotics we primarily train a single robot for a single task. However, modular robot systems now allow for the flexible combination of general-purpose building blocks into task optimized morphologies. However, given the exponentially large number of possible robot morphologies, training a controller for each new design is impractical. In this work, we propose MetaMorph, a Transformer based approach to learn a universal controller over a modular robot design space. MetaMorph is based on the insight that robot morphology is just another modality on which we can condition the output of a Transformer. Through extensive experiments we demonstrate that large scale pre-training on a variety of robot morphologies results in policies with combinatorial generalization capabilities, including zero shot generalization to unseen robot morphologies. We further demonstrate that our pre-trained policy can be used for sample-efficient transfer to completely new robot morphologies and tasks.

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。

↑