Planning Multi-Fingered Grasps as Probabilistic Inference in a Learned Deep Network
Planning Multi-Fingered Grasps as Probabilistic Inference in a Learned Deep Network
复制标题
DOI：
10.1007/978-3-030-28619-4_35
复制
发表时间：
2018-04
期刊：
影响因子：
--
通讯作者：
Qingkai Lu;Kautilya Chenna;Balakumar Sundaralingam;Tucker Hermans
中科院分区：
文献类型：
--
作者：
Qingkai Lu;Kautilya Chenna;Balakumar Sundaralingam;Tucker Hermans
我们提出了一种利用学习的深度神经网络模型进行多指抓取规划的新颖方法。我们训练卷积神经网络来预测抓取成功与否，作为对象视觉信息和抓取配置的函数。然后，我们可以制定抓取规划来推断最大化抓取成功概率的抓取配置。我们使用反向传播算法在神经网络内部使用梯度上升优化来有效地执行此推理。我们的工作是第一个使用深度神经网络在配置空间中直接规划高质量的多指抓取，而不需要外部规划器。我们验证了我们的推理方法，在真实的机器人上执行多指和两指抓取。我们的实验结果表明，我们的规划方法优于现有的神经网络规划方法；同时提供其他一些好处，包括学习中的数据效率以及足够快的速度以部署到真正的机器人应用程序中。
We propose a novel approach to multi-fingered grasp planning leveraging learned deep neural network models. We train a convolutional neural network to predict grasp success as a function of both visual information of an object and grasp configuration. We can then formulate grasp planning as inferring the grasp configuration which maximizes the probability of grasp success. We efficiently perform this inference using a gradient-ascent optimization inside the neural network using the backpropagation algorithm. Our work is the first to directly plan high quality multi-fingered grasps in configuration space using a deep neural network without the need of an external planner. We validate our inference method performing both multi-finger and two-finger grasps on real robots. Our experimental results show that our planning method outperforms existing planning methods for neural networks; while offering several other benefits including being data-efficient in learning and fast enough to be deployed in real robotic applications.