VBM-NET：基于等变TransporterNet与GNN的移动操作视觉基座位姿学习
VBM-NET: Visual Base Pose Learning for Mobile Manipulation using Equivariant TransporterNet and GNNs
AI总结：
针对移动操作中基座位姿规划依赖精确状态信息的问题，提出结合等变TransporterNet与GNN的VBM-NET，基于俯视投影通过强化学习选最优基座位姿，效率优于经典方法且实现仿真到现实的迁移。
摘要：
在移动操作领域，选择最优的移动基座位姿是成功抓取物体的关键。现有研究要么通过经典规划方法，要么通过学习基于状态的策略来解决这一问题，它们均假设能够获取可靠的状态信息，例如精确的物体位姿和环境模型。本文中，我们直接基于场景的俯视正交投影研究基座位姿规划，这类投影既能提供场景全局概览，又能保留空间结构。我们提出VBM-NET，这是一种利用上述俯视正交投影进行基座位姿选择的基于学习的方法。我们采用等变TransporterNet利用空间对称性，高效学习可供抓取的候选基座位姿。此外，我们使用图神经网络（GNNs）表示数量可变的候选基座位姿，并通过强化学习从中选出最优基座位姿。实验表明，VBM-NET能生成与经典方法相当的解决方案，且计算时间大幅缩短。此外，我们将仿真中训练的策略成功部署到真实世界移动操作任务中，验证了仿真到现实的迁移效果。
英文摘要：
In Mobile Manipulation, selecting an optimal mobile base pose is essential for successful object grasping. Previous works have addressed this problem either through classical planning methods or by learning state-based policies. They assume access to reliable state information, such as the precise object poses and environment models. In this work, we study base pose planning directly from top-down orthographic projections of the scene, which provide a global overview of the scene while preserving spatial structure. We propose VBM-NET, a learning-based method for base pose selection using such top-down orthographic projections. We use equivariant TransporterNet to exploit spatial symmetries and efficiently learn candidate base poses for grasping. Further, we use graph neural networks to represent a varying number of candidate base poses and use Reinforcement Learning to determine the optimal base pose among them. We show that VBM-NET can produce comparable solutions to the classical methods in significantly less computation time. Furthermore, we validate sim-to-real transfer by successfully deploying a policy trained in simulation to real-world mobile manipulation.
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。
