Sim-To-Real via Sim-To-Sim: Data-Efficient Robotic Grasping via Randomized-To-Canonical Adaptation Networks

@article{James2018SimToRealVS,
  title={Sim-To-Real via Sim-To-Sim: Data-Efficient Robotic Grasping via Randomized-To-Canonical Adaptation Networks},
  author={Stephen James and Paul Wohlhart and Mrinal Kalakrishnan and Dmitry Kalashnikov and Alex Irpan and Julian Ibarz and Sergey Levine and Raia Hadsell and Konstantinos Bousmalis},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2018},
  pages={12619-12629}
}
Highlight Information
Real world data, especially in the domain of robotics, is notoriously costly to collect. [...] Key Method Our method learns to translate randomized rendered images into their equivalent non-randomized, canonical versions. This in turn allows for real images to also be translated into canonical sim images. We demonstrate the effectiveness of this sim-to-real approach by training a vision-based closed-loop grasping reinforcement learning agent in simulation, and then transferring it to the real world to attain 70…Expand Abstract

Figures, Tables, and Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 44 CITATIONS

Learning to Manipulate Object Collections Using Grounded State Representations

VIEW 6 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Morphology-Agnostic Visual Robotic Control

VIEW 3 EXCERPTS
CITES BACKGROUND & METHODS

Adversarial Imitation Learning between Agents with Different Numbers of State Dimensions

  • Taketo Yoshida, Yasuo Kuniyoshi
  • Computer Science
  • 2019 IEEE Second International Conference on Artificial Intelligence and Knowledge Engineering (AIKE)
  • 2019
VIEW 1 EXCERPT
CITES METHODS

Affordance Learning for End-to-End Visuomotor Robot Control

VIEW 1 EXCERPT
CITES METHODS

References

Publications referenced by this paper.
SHOWING 1-10 OF 72 REFERENCES

Using Simulation and Domain Adaptation to Improve Efficiency of Deep Robotic Grasping

VIEW 4 EXCERPTS

VR-Goggles for Robots: Real-to-Sim Domain Adaptation for Visual Control

VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experience

VIEW 1 EXCERPT

Learning Synergies Between Pushing and Grasping with Self-Supervised Deep Reinforcement Learning

VIEW 1 EXCERPT

Pybullet, a python module for physics simulation for games, robotics and machine learning

  • Erwin Coumans, Yunfei Bai
  • 2018
VIEW 1 EXCERPT