Generative Attention Learning: a “GenerAL” framework for high-performance multi-fingered grasping in clutter
Generative Attention Learning: a “GenerAL” framework for high-performance multi-fingered grasping in clutter
复制标题
DOI：
10.1007/s10514-020-09907-y
复制
发表时间：
2020-02
影响因子：
3.5
通讯作者：
Bohan Wu;Iretiayo Akinola;Abhi Gupta;Feng Xu;Jacob Varley;David Watkins-Valls;P. Allen
中科院分区：
文献类型：
--
作者：
Bohan Wu;Iretiayo Akinola;Abhi Gupta;Feng Xu;Jacob Varley;David Watkins-Valls;P. Allen
生成性注意学习是一种高自由度多指抓取框架，它不仅对密集杂波和新物体具有较强的鲁棒性，而且对各种不同的并颌和多指机器人手也是有效的。该框架引入了一种新颖的注意机制，大大提高了杂波中的抓取成功率。它的生成性允许学习具有灵活末端执行器位置和方向的全自由度抓取，以及手的所有手指关节角度。经过纯粹的模拟训练，这个框架巧妙地缩小了模拟与真实之间的差距。为了缩小视觉模拟与真实之间的差距，该框架使用单个深度图像作为输入。为了缩小动态模拟与真实的差距，该框架绕过了连续的运动控制，从相同的深度图像推断出从像素到笛卡尔空间的直接映射。最后，该框架通过使用两个不同自由度的多指机器人手臂系统，在具有新对象的杂乱场景中实现超现实世界的抓取成功率，从而展示了机器人之间的通用性。
Generative Attention Learning (GenerAL) is a framework for high-DOF multi-fingered grasping that is not only robust to dense clutter and novel objects but also effective with a variety of different parallel-jaw and multi-fingered robot hands. This framework introduces a novel attention mechanism that substantially improves the grasp success rate in clutter. Its generative nature allows the learning of full-DOF grasps with flexible end-effector positions and orientations, as well as all finger joint angles of the hand. Trained purely in simulation, this framework skillfully closes the sim-to-real gap. To close the visual sim-to-real gap, this framework uses a single depth image as input. To close the dynamics sim-to-real gap, this framework circumvents continuous motor control with a direct mapping from pixel to Cartesian space inferred from the same depth image. Finally, this framework demonstrates inter-robot generality by achieving overreal-world grasp success rates in cluttered scenes with novel objects using two multi-fingered robotic hand-arm systems with different degrees of freedom.