English

Scoring Graspability based on Grasp Regression for Better Grasp Prediction

Robotics 2021-04-01 v3 Artificial Intelligence Machine Learning

Abstract

Grasping objects is one of the most important abilities that a robot needs to master in order to interact with its environment. Current state-of-the-art methods rely on deep neural networks trained to jointly predict a graspability score together with a regression of an offset with respect to grasp reference parameters. However, these two predictions are performed independently, which can lead to a decrease in the actual graspability score when applying the predicted offset. Therefore, in this paper, we extend a state-of-the-art neural network with a scorer that evaluates the graspability of a given position, and introduce a novel loss function which correlates regression of grasp parameters with graspability score. We show that this novel architecture improves performance from 82.13% for a state-of-the-art grasp detection network to 85.74% on Jacquard dataset. When the learned model is transferred onto a real robot, the proposed method correlating graspability and grasp regression achieves a 92.4% rate compared to 88.1% for the baseline trained without the correlation.

Keywords

Cite

@article{arxiv.2002.00872,
  title  = {Scoring Graspability based on Grasp Regression for Better Grasp Prediction},
  author = {Amaury Depierre and Emmanuel Dellandréa and Liming Chen},
  journal= {arXiv preprint arXiv:2002.00872},
  year   = {2021}
}
R2 v1 2026-06-23T13:29:31.631Z