Learning Continuous 3D Reconstructions for Geometrically Aware Grasping
Learning Continuous 3D Reconstructions for Geometrically Aware Grasping
复制标题
DOI：
10.1109/icra40945.2020.9196981
复制
发表时间：
2019-10
期刊：
影响因子：
--
通讯作者：
Mark Van der Merwe;Qingkai Lu;Balakumar Sundaralingam;Martin Matak;Tucker Hermans
中科院分区：
文献类型：
--
作者：
Mark Van der Merwe;Qingkai Lu;Balakumar Sundaralingam;Martin Matak;Tucker Hermans
深度学习使得从部分对象视图中抓取以前看不见的对象的抓取合成有了显着的改进。然而，现有的方法缺乏能力，明确的原因，在选择一个把握的对象的完整的3D几何形状，依赖于间接的几何推理时，学习把握成功的网络。这放弃了显式几何推理，例如避免不期望的机器人对象碰撞。我们建议利用一种新的，学习3D重建，使几何意识在抓持系统。我们利用重建网络的结构来学习抓取成功分类器，该分类器作为连续抓取优化的目标函数。我们还明确地约束优化，以避免不必要的接触，直接使用重建。我们研究的作用，几何形状在把握指标的培训，并通过96个机器人把握审判。我们的结果可以在https://sites.google.com/view/reconstruction-grasp/上找到。
Deep learning has enabled remarkable improvements in grasp synthesis for previously unseen objects from partial object views. However, existing approaches lack the ability to explicitly reason about the full 3D geometry of the object when selecting a grasp, relying on indirect geometric reasoning derived when learning grasp success networks. This abandons explicit geometric reasoning, such as avoiding undesired robot object collisions. We propose to utilize a novel, learned 3D reconstruction to enable geometric awareness in a grasping system. We leverage the structure of the reconstruction network to learn a grasp success classifier which serves as the objective function for a continuous grasp optimization. We additionally explicitly constrain the optimization to avoid undesired contact, directly using the reconstruction. We examine the role of geometry in grasping both in the training of grasp metrics and through 96 robot grasping trials. Our results can be found on https://sites.google.com/view/reconstruction-grasp/.