等变描述子场：用于端到端视觉机器人操作学习的SE(3)-等变能量基模型
Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learning
AI总结：
针对视觉机器人操作端到端学习样本效率低的问题，提出基于李群表示理论的SE(3)-等变能量基模型，仅需5~10次演示即可从零学习，且在新位姿、新实例、有干扰的任务中泛化性良好。
摘要：
众所周知，视觉机器人操作的端到端学习存在样本效率低下的问题，需要大量演示数据。空间旋转平移等变性，即SE(3)-等变性，可用于提升机器人操作学习的样本效率。本文提出了可完全端到端训练的、基于点云的视觉机器人操作SE(3)-等变模型。利用李群表示理论，我们构建了新型SE(3)-等变能量基模型，可实现高样本效率的端到端学习。研究表明，该模型无需先验知识即可从零开始学习，且样本效率极高（仅需5~10次演示即可）。此外，模型可泛化至具备以下特征的任务：（i）前所未见的目标物体位姿、（ii）同类别前所未见的目标物体实例、（iii）前所未见的视觉干扰物。我们通过6自由度机器人操作任务验证了模型的样本效率与泛化能力。代码可在以下链接获取：https://github.com/tomato1mule/edf
英文摘要：
End-to-end learning for visual robotic manipulation is known to suffer from sample inefficiency, requiring large numbers of demonstrations. The spatial roto-translation equivariance, or the SE(3)-equivariance can be exploited to improve the sample efficiency for learning robotic manipulation. In this paper, we present SE(3)-equivariant models for visual robotic manipulation from point clouds that can be trained fully end-to-end. By utilizing the representation theory of the Lie group, we construct novel SE(3)-equivariant energy-based models that allow highly sample efficient end-to-end learning. We show that our models can learn from scratch without prior knowledge and yet are highly sample efficient (5~10 demonstrations are enough). Furthermore, we show that our models can generalize to tasks with (i) previously unseen target object poses, (ii) previously unseen target object instances of the category, and (iii) previously unseen visual distractors. We experiment with 6-DoF robotic manipulation tasks to validate our models' sample efficiency and generalizability. Codes are available at: https://github.com/tomato1mule/edf
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。
