TY - GEN
T1 - The CoSTAR Block Stacking Dataset
T2 - 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2019
AU - Hundt, Andrew
AU - Jain, Varun
AU - Lin, Chia Hung
AU - Paxton, Chris
AU - Hager, Gregory D.
N1 - Funding Information:
VII. ACKNOWLEDGEMENTS We thank Chunting Jiao for his assistance with data collection. This material is based upon work supported by the National Science Foundation under NSF NRI Grant Award No. 1637949.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/11
Y1 - 2019/11
N2 - A robot can now grasp an object more effectively than ever before, but once it has the object what happens next? We show that a mild relaxation of the task and workspace constraints implicit in existing object grasping datasets can cause neural network based grasping algorithms to fail on even a simple block stacking task when executed under more realistic circumstances. To address this, we introduce the JHU CoSTAR Block Stacking Dataset (BSD), where a robot interacts with 5.1 cm colored blocks to complete an order-fulfillment style block stacking task. It contains dynamic scenes and real time-series data in a less constrained environment than comparable datasets. There are nearly 12,000 stacking attempts and over 2 million frames of real data. We discuss the ways in which this dataset provides a valuable resource for a broad range of other topics of investigation. We find that hand-designed neural networks that work on prior datasets do not generalize to this task. Thus, to establish a baseline for this dataset, we demonstrate an automated search of neural network based models using a novel multiple-input HyperTree MetaModel, and find a final model which makes reasonable 3D pose predictions for grasping and stacking on our dataset. The CoSTAR BSD, code, and instructions are available at sites.google.com/site/costardataset.
AB - A robot can now grasp an object more effectively than ever before, but once it has the object what happens next? We show that a mild relaxation of the task and workspace constraints implicit in existing object grasping datasets can cause neural network based grasping algorithms to fail on even a simple block stacking task when executed under more realistic circumstances. To address this, we introduce the JHU CoSTAR Block Stacking Dataset (BSD), where a robot interacts with 5.1 cm colored blocks to complete an order-fulfillment style block stacking task. It contains dynamic scenes and real time-series data in a less constrained environment than comparable datasets. There are nearly 12,000 stacking attempts and over 2 million frames of real data. We discuss the ways in which this dataset provides a valuable resource for a broad range of other topics of investigation. We find that hand-designed neural networks that work on prior datasets do not generalize to this task. Thus, to establish a baseline for this dataset, we demonstrate an automated search of neural network based models using a novel multiple-input HyperTree MetaModel, and find a final model which makes reasonable 3D pose predictions for grasping and stacking on our dataset. The CoSTAR BSD, code, and instructions are available at sites.google.com/site/costardataset.
UR - http://www.scopus.com/inward/record.url?scp=85081160051&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85081160051&partnerID=8YFLogxK
U2 - 10.1109/IROS40897.2019.8967784
DO - 10.1109/IROS40897.2019.8967784
M3 - Conference contribution
AN - SCOPUS:85081160051
T3 - IEEE International Conference on Intelligent Robots and Systems
SP - 1797
EP - 1804
BT - 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2019
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 3 November 2019 through 8 November 2019
ER -