arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2210.03370cs.ROcs.AIcs.LG
收藏

GNM:驱动任意机器人的通用导航模型

GNM: A General Navigation Model to Drive Any Robot

UC Berkeley(加州大学伯克利分校);Toyota Motor North America(丰田北美汽车公司)

Dhruv Shah, Ajay Sridhar, Arjun Bhorkar, Noriaki Hirose, Sergey Levine

AI总结:

本文提出通用导航模型GNM,通过在6种机器人的60小时异构导航数据上训练全能策略,结合时间上下文与标准化动作空间,实现了跨环境与具身形态的广泛泛化,显著提升了导航鲁棒性。

摘要:

学习为基于视觉的导航提供了强大的工具,但基于学习的策略能力受限于有限的训练数据。如果能结合来自所有可用来源的数据,包括多种机器人,我们就能训练出更强大的导航模型。本文研究如何在从许多不同但结构相似的机器人获取的数据上,训练一个用于基于视觉导航的通用目标条件模型,并实现跨环境和跨具身形态的广泛泛化。我们分析了跨机器人有效数据共享所需的设计决策,包括时间上下文和标准化动作空间的使用,并证明了从异构数据集训练出的全能策略优于在任何单一数据集上训练的策略。我们整理了来自6种不同机器人的60小时导航轨迹,并将训练好的GNM部署到一系列新机器人上,包括一个欠驱动四旋翼飞行器。我们发现,在多样化数据上训练能带来针对感知和驱动退化情况的鲁棒性。使用具有广泛泛化能力的预训练导航模型可以为未来在新型机器人上的应用提供引导,我们希望GNM能代表朝着这一方向迈出的一步。有关数据集、代码和视频的更多信息,请查看我们的项目页面 https://sites.google.com/view/drive-any-robot。

英文摘要:

Learning provides a powerful tool for vision-based navigation, but the capabilities of learning-based policies are constrained by limited training data. If we could combine data from all available sources, including multiple kinds of robots, we could train more powerful navigation models. In this paper, we study how a general goal-conditioned model for vision-based navigation can be trained on data obtained from many distinct but structurally similar robots, and enable broad generalization across environments and embodiments. We analyze the necessary design decisions for effective data sharing across robots, including the use of temporal context and standardized action spaces, and demonstrate that an omnipolicy trained from heterogeneous datasets outperforms policies trained on any single dataset. We curate 60 hours of navigation trajectories from 6 distinct robots, and deploy the trained GNM on a range of new robots, including an underactuated quadrotor. We find that training on diverse data leads to robustness against degradation in sensing and actuation. Using a pre-trained navigation model with broad generalization capabilities can bootstrap applications on novel robots going forward, and we hope that the GNM represents a step in that direction. For more information on the datasets, code, and videos, please check out our project page https://sites.google.com/view/drive-any-robot.

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。

↑