arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.04576cs.ROcs.AI
收藏

ActionFlow：利用空间对称 Flow Matching 实现等变、准确且高效的策略

ActionFlow: Equivariant, Accurate, and Efficient Policies with Spatially Symmetric Flow Matching

TU Darmstadt（达姆施塔特工业大学）；German Research Center for AI (DFKI)（德国人工智能研究中心）

Niklas Funk, Julen Urain, Joao Carvalho, Vignesh Prasad, Georgia Chalvatzaki, Jan Peters

阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出 ActionFlow 策略框架，通过 SE(3) 不变 Transformer 与 Flow Matching 结合空间对称归纳偏置，实现机器人操作中等变、准确、高效的动作生成。

摘要：

空间理解是大多数机器人任务的关键方面，尤其是在泛化能力很重要时。尽管深度生成模型已在复杂操作任务中取得令人瞩目的结果，但由于缺乏一种能够编码观测与动作之间复杂空间关系的表示，空间泛化往往受到限制，从而需要大量演示。为解决这一问题，我们提出了一种新的策略类别 ActionFlow。ActionFlow 在生成富有表达能力的动作序列的同时，融入了空间对称性归纳偏置。在表示层面，ActionFlow 引入了 SE(3) 不变 Transformer 架构，能够基于观测与动作之间的相对 SE(3) 位姿进行有依据的空间推理。在动作生成方面，ActionFlow 利用 Flow Matching，这是一种先进的深度生成模型，以生成高质量样本和快速推理著称，而快速推理正是反馈控制的重要属性。综合来看，ActionFlow 策略表现出强烈的空间偏置和局部性偏置，以及 SE(3) 等变的动作生成。我们的实验在多个仿真和真实世界机器人操作任务上证明了 ActionFlow 及其两个主要组成部分的有效性，并证实我们能够通过空间对称的 flow matching 获得等变、准确且高效的策略。项目网站：https://flowbasedpolicies.github.io/

英文摘要：

Spatial understanding is a critical aspect of most robotic tasks, particularly when generalization is important. Despite the impressive results of deep generative models in complex manipulation tasks, the absence of a representation that encodes intricate spatial relationships between observations and actions often limits spatial generalization, necessitating large amounts of demonstrations. To tackle this problem, we introduce a novel policy class, ActionFlow. ActionFlow integrates spatial symmetry inductive biases while generating expressive action sequences. On the representation level, ActionFlow introduces an SE(3) Invariant Transformer architecture, which enables informed spatial reasoning based on the relative SE(3) poses between observations and actions. For action generation, ActionFlow leverages Flow Matching, a state-of-the-art deep generative model known for generating high-quality samples with fast inference - an essential property for feedback control. In combination, ActionFlow policies exhibit strong spatial and locality biases and SE(3)-equivariant action generation. Our experiments demonstrate the effectiveness of ActionFlow and its two main components on several simulated and real-world robotic manipulation tasks and confirm that we can obtain equivariant, accurate, and efficient policies with spatially symmetric flow matching. Project website: https://flowbasedpolicies.github.io/

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。

↑