当前位置: X-MOL 学术Auton. Robot. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
VisuoSpatial Foresight for physical sequential fabric manipulation
Autonomous Robots ( IF 3.5 ) Pub Date : 2021-07-26 , DOI: 10.1007/s10514-021-10001-0
Ryan Hoque 1 , Daniel Seita 1 , Ashwin Balakrishna 1 , Aditya Ganapathi 1 , Ajay Kumar Tanwani 1 , Ken Goldberg 1 , Nawid Jamali 2 , Katsu Yamane 2 , Soshi Iba 2
Affiliation  

Robotic fabric manipulation has applications in home robotics, textiles, senior care and surgery. Existing fabric manipulation techniques, however, are designed for specific tasks, making it difficult to generalize across different but related tasks. We build upon the Visual Foresight framework to learn fabric dynamics that can be efficiently reused to accomplish different sequential fabric manipulation tasks with a single goal-conditioned policy. We extend our earlier work on VisuoSpatial Foresight (VSF), which learns visual dynamics on domain randomized RGB images and depth maps simultaneously and completely in simulation. In this earlier work, we evaluated VSF on multi-step fabric smoothing and folding tasks against 5 baseline methods in simulation and on the da Vinci Research Kit surgical robot without any demonstrations at train or test time. A key finding was that depth sensing significantly improves performance: RGBD data yields an \(\mathbf{80 \%}\) improvement in fabric folding success rate in simulation over pure RGB data. In this work, we vary 4 components of VSF, including data generation, visual dynamics model, cost function, and optimization procedure. Results suggest that training visual dynamics models using longer, corner-based actions can improve the efficiency of fabric folding by 76% and enable a physical sequential fabric folding task that VSF could not previously perform with 90% reliability. Code, data, videos, and supplementary material are available at https://sites.google.com/view/fabric-vsf/.



中文翻译:

用于物理顺序结构操作的 VisuoSpatial Foresight

机器人织物操作已应用于家庭机器人、纺织品、高级护理和外科手术。然而,现有的织物操作技术是为特定任务而设计的,因此很难概括不同但相关的任务。我们建立在 Visual Foresight 框架的基础上来学习结构动力学,这些结构动力学可以有效地重用,以使用单一目标条件策略完成不同的顺序结构操作任务。我们扩展了我们早期在 VisuoSpatial Foresight (VSF) 上的工作,它在模拟中同时完全地学习域随机 RGB 图像和深度图上的视觉动态。在早期的这项工作中,我们针对模拟中的 5 种基线方法和 da Vinci Research Kit 手术机器人在多步织物平滑和折叠任务上评估了 VSF,而没有在训练或测试时间进行任何演示。\(\mathbf{80 \%}\)在模拟纯 RGB 数据时提高织物折叠成功率。在这项工作中,我们改变了 VSF 的 4 个组件,包括数据生成、视觉动力学模型、成本函数和优化程序。结果表明,使用更长的、基于角的动作训练视觉动力学模型可以将织物折叠的效率提高 76%,并使 VSF 以前无法以 90% 的可靠性执行物理顺序织物折叠任务。代码、数据、视频和补充材料可从 https://sites.google.com/view/fabric-vsf/ 获得。

更新日期:2021-07-26
down
wechat
bug