REGRAD: A Large-Scale Relational Grasp Dataset for Safe and Object-Specific Robotic Grasping in Clutter

Despite the impressive progress achieved in robotic grasping, robots are not skilled in sophisticated tasks (e.g. search and grasp a specified target in clutter). Such tasks involve not only grasping but the comprehensive perception of the world (e.g. the object relationships). Recently, encouraging...

Full description

Saved in:
Bibliographic Details
Published inIEEE robotics and automation letters Vol. 7; no. 2; pp. 2929 - 2936
Main Authors Zhang, Hanbo, Yang, Deyu, Wang, Han, Zhao, Binglei, Lan, Xuguang, Ding, Jishiyu, Zheng, Nanning
Format Journal Article
LanguageEnglish
Published Piscataway IEEE 01.04.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Algorithms
Annotations
Clutter
Datasets
deep learning
Grasping
Grasping (robotics)
Image segmentation
Machine learning
manipulation relationship
perception for manipulation
Performance evaluation
Point cloud compression
robot vision
Robotic grasping
Robots
Task analysis
Three dimensional models
Visualization
Online AccessGet full text

Cover

More Information
Summary:Despite the impressive progress achieved in robotic grasping, robots are not skilled in sophisticated tasks (e.g. search and grasp a specified target in clutter). Such tasks involve not only grasping but the comprehensive perception of the world (e.g. the object relationships). Recently, encouraging results demonstrate that it is possible to understand high-level concepts by learning. However, such algorithms are usually data-intensive, and the lack of data severely limits their performance. In this letter, we present a new dataset named REGRAD for the learning of relationships among objects and grasps. We collect the annotations of object poses, segmentations, grasps, and relationships for the target-driven relational grasping tasks. Our dataset is collected in both forms of 2D images and 3D point clouds. Moreover, since all the data are generated automatically, it is free to import new objects for data generation. We also released a real-world validation dataset to evaluate the sim-to-real performance of models trained on REGRAD. Finally, we conducted a series of experiments, showing that the models trained on REGRAD could generalize well to the realistic scenarios, in terms of both relationship and grasp detection. Our dataset and code could be found at. 1
ISSN:2377-3766
2377-3766
DOI:10.1109/LRA.2022.3142401