Sample Efficient Grasp Learning Using Equivariant Models
Sample Efficient Grasp Learning Using Equivariant Models
复制标题
DOI：
10.15607/rss.2022.xviii.071
复制
发表时间：
2022-02
期刊：
影响因子：
--
通讯作者：
Xu Zhu;Dian Wang;Ondrej Biza;Guanang Su;R. Walters;Robert W. Platt
中科院分区：
文献类型：
--
作者：
Xu Zhu;Dian Wang;Ondrej Biza;Guanang Su;R. Walters;Robert W. Platt
在平面抓取检测中，目标是从场景的图像中学习一个函数到$\mathrm{SE}（2）$中的一组可行抓取姿势。在本文中，我们认识到最佳抓取函数是$\mathrm{SE}（2）$-等变的，可以使用等变卷积神经网络建模。因此，我们能够显着提高抓取学习的样本效率，仅在600次抓取尝试后就获得了抓取函数的良好近似。这是足够少的，我们可以在大约1.5小时内学会完全掌握物理机器人。
In planar grasp detection, the goal is to learn a function from an image of a scene onto a set of feasible grasp poses in $\mathrm{SE}(2)$. In this paper, we recognize that the optimal grasp function is $\mathrm{SE}(2)$-equivariant and can be modeled using an equivariant convolutional neural network. As a result, we are able to significantly improve the sample efficiency of grasp learning, obtaining a good approximation of the grasp function after only 600 grasp attempts. This is few enough that we can learn to grasp completely on a physical robot in about 1.5 hours.