Learning task-oriented grasping for tool manipulation from simulated self-supervision
Learning task-oriented grasping for tool manipulation from simulated self-supervision
复制标题
DOI：
10.1177/0278364919872545
复制
发表时间：
2019-08-29
影响因子：
9.2
通讯作者：
Savarese, Silvio
中科院分区：
文献类型：
--
作者：
Kuan Fang;Zhu, Yuke;Savarese, Silvio
工具操作对于帮助机器人完成具有挑战性的任务目标至关重要。它需要对任务的预期效果进行推理，从而正确地掌握和操纵工具来完成任务。机器人领域的大多数工作都集中在任务不可知的抓取上，它只优化抓取鲁棒性，而不考虑后续的操作任务。在这篇文章中，我们提出了面向任务的抓取网络（TOG-Net），以共同优化面向任务的工具和该工具的操作策略。该模型的训练过程是基于大规模的模拟自我监督与程序生成的工具对象。我们进行模拟和现实世界的实验两个工具为基础的操作任务：清扫和锤击。我们的模型实现了71.1%的清扫任务成功率和80.0%的锤击任务成功率。
Tool manipulation is vital for facilitating robots to complete challenging task goals. It requires reasoning about the desired effect of the task and, thus, properly grasping and manipulating the tool to achieve the task. Most work in robotics has focused on task-agnostic grasping, which optimizes for only grasp robustness without considering the subsequent manipulation tasks. In this article, we propose the Task-Oriented Grasping Network (TOG-Net) to jointly optimize both task-oriented grasping of a tool and the manipulation policy for that tool. The training process of the model is based on large-scale simulated self-supervision with procedurally generated tool objects. We perform both simulated and real-world experiments on two tool-based manipulation tasks: sweeping and hammering. Our model achieves overall 71.1% task success rate for sweeping and 80.0% task success rate for hammering.