Deep Imitation Learning of Sequential Fabric Smoothing From an Algorithmic Supervisor
Deep Imitation Learning of Sequential Fabric Smoothing From an Algorithmic Supervisor
复制标题
来自算法主管的顺序结构平滑的深度模仿学习
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Ken Goldberg
中科院分区：
文献类型：
--
作者：
Daniel Seita;Aditya Ganapathi;Ryan Hoque;M. Hwang;Edward Cen;A. Tanwani;A. Balakrishna;Brijen Thananjeyan;Jeffrey Ichnowski;Nawid Jamali;K. Yamane;Soshi Iba;J. Canny;Ken Goldberg
用于展平和抚平织物的顺序牵拉策略在从手术到制造业再到诸如铺床和叠衣服等家庭任务中都有应用。由于织物状态和动力学的复杂性，我们应用深度模仿学习来学习策略，即给定矩形织物样本的颜色（RGB）、深度（D）或颜色 - 深度组合（RGBD）图像，估计抓取点和牵拉向量以展开织物从而使覆盖面积最大化。为了生成数据，我们开发了一个织物模拟器和一个能够获取完整状态信息的算法监督器。我们在模拟中使用领域随机化和数据集聚合（DAgger）对初始随机配置中的三个难度层级训练策略。我们展示了将五种基线策略与学习到的策略进行比较的结果，并报告了将RGB、D和RGBD图像作为输入的系统比较。在模拟中，学习到的策略实现了与分析基线相当或更优的性能。在使用达芬奇研究工具包（dVRK）手术机器人进行的180次物理实验中，在模拟中训练的RGBD策略根据难度层级实现了83%到95%的覆盖率，这表明可以从算法监督器中学习到有效的织物抚平策略，并且深度传感是对仅颜色信息的一种有价值的补充。补充材料可在https://sites.google.com/view/fabric - smoothing获取。
Sequential pulling policies to flatten and smooth fabrics have applications from surgery to manufacturing to home tasks such as bed making and folding clothes. Due to the complexity of fabric states and dynamics, we apply deep imitation learning to learn policies that, given color (RGB), depth (D), or combined color-depth (RGBD) images of a rectangular fabric sample, estimate pick points and pull vectors to spread the fabric to maximize coverage. To generate data, we develop a fabric simulator and an algorithmic supervisor that has access to complete state information. We train policies in simulation using domain randomization and dataset aggregation (DAgger) on three tiers of difficulty in the initial randomized configuration. We present results comparing five baseline policies to learned policies and report systematic comparisons of RGB vs D vs RGBD images as inputs. In simulation, learned policies achieve comparable or superior performance to analytic baselines. In 180 physical experiments with the da Vinci Research Kit (dVRK) surgical robot, RGBD policies trained in simulation attain coverage of 83% to 95% depending on difficulty tier, suggesting that effective fabric smoothing policies can be learned from an algorithmic supervisor and that depth sensing is a valuable addition to color alone. Supplementary material is available at https://sites.google.com/view/fabric-smoothing.
DOI：
--
发表时间：
2018-06
期刊：
ArXiv
影响因子：
--
作者：
J. Matas;Stephen James;A. Davison
通讯作者：
J. Matas;Stephen James;A. Davison