In this paper, we propose a novel representation for grasping using contacts between multi-finger robotic hands and objects to be manipulated. This representation significantly reduces the prediction dimensions and accelerates the learning process. We present an effective end-to-end network, CMG-Net, for grasping unknown objects in a cluttered environment by efficiently predicting multi-finger grasp poses and hand configurations from a single-shot point cloud. Moreover, we create a synthetic grasp dataset that consists of five thousand cluttered scenes, 80 object categories, and 20 million annotations. We perform a comprehensive empirical study and demonstrate the effectiveness of our grasping representation and CMG-Net. Our work significantly outperforms the state-of-the-art for three-finger robotic hands. We also demonstrate that the model trained using synthetic data performs very well for real robots.
@article{arxiv.2303.13182,
title = {CMG-Net: An End-to-End Contact-Based Multi-Finger Dexterous Grasping Network},
author = {Mingze Wei and Yaomin Huang and Zhiyuan Xu and Ning Liu and Zhengping Che and Xinyu Zhang and Chaomin Shen and Feifei Feng and Chun Shan and Jian Tang},
journal= {arXiv preprint arXiv:2303.13182},
year = {2023}
}
Comments
The first two authors are with equal contributions. Paper accepted by ICRA 2023