English

Improved GQ-CNN: Deep Learning Model for Planning Robust Grasps

Machine Learning 2018-02-19 v1 Artificial Intelligence Robotics Machine Learning

Abstract

Recent developments in the field of robot grasping have shown great improvements in the grasp success rates when dealing with unknown objects. In this work we improve on one of the most promising approaches, the Grasp Quality Convolutional Neural Network (GQ-CNN) trained on the DexNet 2.0 dataset. We propose a new architecture for the GQ-CNN and describe practical improvements that increase the model validation accuracy from 92.2% to 95.8% and from 85.9% to 88.0% on respectively image-wise and object-wise training and validation splits.

Keywords

Cite

@article{arxiv.1802.05992,
  title  = {Improved GQ-CNN: Deep Learning Model for Planning Robust Grasps},
  author = {Maciej Jaśkowski and Jakub Świątkowski and Michał Zając and Maciej Klimek and Jarek Potiuk and Piotr Rybicki and Piotr Polatowski and Przemysław Walczyk and Kacper Nowicki and Marek Cygan},
  journal= {arXiv preprint arXiv:1802.05992},
  year   = {2018}
}

Comments

6 pages, 3 figures

R2 v1 2026-06-23T00:24:41.816Z