Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours

Lerrel Pinto, Abhinav Gupta

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Current model free learning-based robot grasping approaches exploit human-labeled datasets for training the models. However, there are two problems with such a methodology: (a) since each object can be grasped in multiple ways, manually labeling grasp locations is not a trivial task; (b) human labeling is biased by semantics. While there have been attempts to train robots using trial-and-error experiments, the amount of data used in such experiments remains substantially low and hence makes the learner prone to over-fitting. In this paper, we take the leap of increasing the available training data to 40 times more than prior work, leading to a dataset size of 50K data points collected over 700 hours of robot grasping attempts. This allows us to train a Convolutional Neural Network (CNN) for the task of predicting grasp locations without severe overfitting. In our formulation, we recast the regression problem to an 18-way binary classification over image patches. We also present a multi-stage learning approach where a CNN trained in one stage is used to collect hard negatives in subsequent stages. Our experiments clearly show the benefit of using large-scale datasets (and multi-stage training) for the task of grasping. We also compare to several baselines and show state-of-the-art performance on generalization to unseen objects for grasping.

Original languageEnglish (US)
Title of host publication2016 IEEE International Conference on Robotics and Automation, ICRA 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3406-3413
Number of pages8
ISBN (Electronic)9781467380263
DOIs
StatePublished - Jun 8 2016
Event2016 IEEE International Conference on Robotics and Automation, ICRA 2016 - Stockholm, Sweden
Duration: May 16 2016May 21 2016

Publication series

NameProceedings - IEEE International Conference on Robotics and Automation
Volume2016-June
ISSN (Print)1050-4729

Other

Other2016 IEEE International Conference on Robotics and Automation, ICRA 2016
CountrySweden
CityStockholm
Period5/16/165/21/16

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours'. Together they form a unique fingerprint.

Cite this