2019

TossingBot: Learning to Throw Arbitrary Objects with Residual Physics

Zeng, Andy, Song, Shuran, Lee, Johnny et al.

Understand

We investigate whether a robot arm can learn to pick and throw arbitrary objects into selected boxes quickly and accurately.

Throwing has the potential to increase the physical reachability and picking speed of a robot arm.
However, precisely throwing arbitrary objects in unstructured settings presents many challenges: from acquiring reliable pre-throw conditions (e.g.
initial pose of object in manipulator) to handling varying object-centric properties (e.g.

Built on

Task-level robot learning
Eric W Aboaf, Christopher G Atkeson, and David J Reinkensmeyer · 1988
Earlier work this paper cites.
Dynamic manipulation
Matthew T Mason and Kevin M Lynch · 1993
Earlier work this paper cites.
A volumetric method for building complex models from range images
Brian Curless and Marc Levoy · 1996
Earlier work this paper cites.
Dynamic nonprehensile manipulation: Controllability, planning, and experiments
Kevin M Lynch and Matthew T Mason · 1999
Earlier work this paper cites.
Using inaccurate models in reinforcement learning
Pieter Abbeel, Morgan Quigley, and Andrew Y Ng · 2006
Earlier work this paper cites.
High-speed throwing motion based on kinetic chain approach
Taku Senoo, Akio Namiki, and Masatoshi Ishikawa · 2008
Earlier work this paper cites.
1-dof dynamic pitching robot that independently controls velocity, angular velocity, and direction of a ball: Contact models and motion planning
Wataru Mori, Jun Ueda, and Tsukasa Ogasawara · 2009
Earlier work this paper cites.
Understanding the difficulty of training deep feedforward neural networks
Xavier Glorot and Yoshua Bengio · 2010
Earlier work this paper cites.
A ball-throwing robot with visual feedback
Jwu-Sheng Hu, Ming-Chih Chien, Yung-Jung Chang, Shyh-Haur Su, and Chen-Yu Kai · 2010
Earlier work this paper cites.
Reinforcement learning to adjust robot movements to new situations
Jens Kober, Erhan Öztop, and Jan Peters · 2011
Earlier work this paper cites.
Kinectfusion: Real-time dense surface mapping and tracking
Richard A Newcombe, Shahram Izadi, Otmar Hilliges, David Molyneaux, David Kim, Andrew J Davison, Pushmeet Kohi, Jamie Shotton, Steve Hodges, and Andrew Fitzgibbon · 2011
Earlier work this paper cites.
Motion control of a ball throwing robot with a flexible robotic arm
Yizhi Gai, Yukinori Kobayashi, Yohei Hoshino, and Takanori Emaru · 2013
Earlier work this paper cites.
Learning task error models for manipulation
Peter Pastor, Mrinal Kalakrishnan, Jonathan Binney, Jonathan Kelly, Ludovic Righetti, Gaurav Sukhatme, and Stefan Schaal · 2013
Earlier work this paper cites.
Extrinsic dexterity: In-hand manipulation with external forces
Nikhil Chaval-Dafle, Alberto Rodriguez, R. Paolini, B. Tang, S. S. Srinivasa, M. Erdmann, M. T. Mason, I. Lundberg, H. Staab, and T. Fuhlbrigge · 2014
Earlier work this paper cites.

Similar

Fast r-cnn
Ross Girshick · 2015
Cited alongside, same era.
Fully convolutional networks for semantic segmentation
Jonathan Long, Evan Shelhamer, and Trevor Darrell · 2015
Cited alongside, same era.
Deep residual learning for image recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun · 2016
Cited alongside, same era.
Prioritized experience replay
Tom Schaul, John Quan, Ioannis Antonoglou, and David Silver · 2016
Cited alongside, same era.
Segnet: A deep convolutional encoder-decoder architecture for image segmentation
Vijay Badrinarayanan, Alex Kendall, and Roberto Cipolla · 2017
Cited alongside, same era.
Learning data-efficient rigid-body contact models: Case study of planar impact
Nima Fazeli, Samuel Zapolsky, Evan Drumwright, and Alberto Rodriguez · 2017
Cited alongside, same era.
Deep predictive policy training using reinforcement learning
Ali Ghadirzadeh, Atsuto Maki, Danica Kragic, and Mårten Björkman · 2017
Cited alongside, same era.
Adapting learned robotics behaviours through policy adjustment
Juan Camilo Gamboa Higuera, David Meger, and Gregory Dudek · 2017
Cited alongside, same era.
Dex-net 2.0: Deep learning to plan robust grasps with synthetic point clouds and analytic grasp metrics
Jeffrey Mahler, Jacky Liang, Sherdil Niyaz, Michael Laskey, Richard Doan, Xinyu Liu, Juan Aparicio Ojea, and Ken Goldberg · 2017
Cited alongside, same era.
3dmatch: Learning local geometric descriptors from rgb-d reconstructions
Andy Zeng, Shuran Song, Matthias Nießner, Matthew Fisher, Jianxiong Xiao, and Thomas Funkhouser · 2017
Cited alongside, same era.
Augmenting physical simulators with stochastic neural networks: Case study of planar pushing and bouncing
Anurag Ajay, Jiajun Wu, Nima Fazeli, Maria Bauza, Leslie P Kaelbling, Joshua B Tenenbaum, and Alberto Rodriguez · 2018
Cited alongside, same era.
Pybullet, a python module for physics simulation for games, robotics and machine learning
Erwin Coumans and Yunfei Bai · 2018
Cited alongside, same era.

Then

Combining learned and analytical models for predicting action effects
Alina Kloss, Stefan Schaal, and Jeannette Bohg · 2018
Later among the works it cites.
Cartman: The low-cost cartesian manipulator that won the amazon robotics challenge
Douglas Morrison, Adam W Tow, M McTaggart, R Smith, N Kelly-Boxall, S Wade-McCue, J Erskine, R Grinover, A Gurman, T Hunn, et al · 2018
Later among the works it cites.
On-policy dataset synthesis for learning deep robot grasping policies based on fully-convolutional grasp quality neural networks
Vishal Satish, Jeffrey Mahler, and Ken Goldberg · 2018
Later among the works it cites.
Residual policy learning
Tom Silver, Kelsey Allen, Josh Tenenbaum, and Leslie Kaelbling · 2018
Later among the works it cites.
Residual reinforcement learning for robot control
Tobias Johannink, Shikhar Bahl, Ashvin Nair, Jianlan Luo, Avinash Kumar, Matthias Loskyll, Juan Aparicio Ojea, Eugen Solowjow, and Sergey Levine · 2019
Closest in time.
Learning ambidextrous robot grasping policies
Jeffrey Mahler, Matthew Matl, Vishal Satish, Michael Danielczuk, Bill DeRose, Stephen McKinley, and Ken Goldberg · 2019
Closest in time.
Optimal shape and motion planning for dynamic planar manipulation
Orion Taylor and Alberto Rodriguez · 2019
Closest in time.
Densephysnet: Learning dense physical object representations via multi-step dynamic interactions
Zhenjia Xu, Jiajun Wu, Andy Zeng, Joshua B Tenenbaum, and Shuran Song · 2019
Closest in time.
Learning Visual Affordances for Robotic Manipulation
Andy Zeng · 2019
Closest in time.
Tossingbot: Learning to throw arbitrary objects with residual physics
Andy Zeng, Shuran Song, Johnny Lee, Alberto Rodriguez, and Thomas Funkhouser · 2019
Closest in time.
Cleargrasp: 3d shape estimation of transparent objects for manipulation
Shreeyak S Sajjan, Matthew Moore, Mike Pan, Ganesh Nagaraja, Johnny Lee, Andy Zeng, and Shuran Song · 2020
Closest in time.
Learning to see before learning to act: Visual pre-training for manipulation
Lin Yen-Chen, Shuran Zeng, Andy Song, Phillip Isola, and Tsung-Yi Lin · 2020
Closest in time.
Form2fit: Learning shape priors for generalizable assembly from disassembly
Kevin Zakka, Andy Zeng, Johnny Lee, and Shuran Song · 2020
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…