Rearrangement with Nonprehensile Manipulation Using Deep Reinforcement Learning

@article{Yuan2018RearrangementWN,
  title={Rearrangement with Nonprehensile Manipulation Using Deep Reinforcement Learning},
  author={Weihao Yuan and Johannes A. Stork and Danica Kragic and Michael Yu Wang and Kaiyu Hang},
  journal={2018 IEEE International Conference on Robotics and Automation (ICRA)},
  year={2018},
  pages={270-277}
}
Rearranging objects on a tabletop surface by means of nonprehensile manipulation is a task which requires skillful interaction with the physical world. Usually, this is achieved by precisely modeling physical properties of the objects, robot, and the environment for explicit planning. In contrast, as explicitly modeling the physical environment is not always feasible and involves various uncertainties, we learn a nonprehensile rearrangement strategy with deep reinforcement learning based on… CONTINUE READING
8
Twitter Mentions

Figures, Tables, Results, and Topics from this paper.

Key Quantitative Results

  • We demonstrate empirical evidence from simulation that our method leads to a success rate of 85%, show that our system can cope with sudden changes of the environment, and compare our performance with human level performance.
  • Finally, we achieve a success rate of 85% indicating that the learned network can effectively handle the task of nonprehensile rearrangement.

Citations

Publications citing this paper.
SHOWING 1-4 OF 4 CITATIONS

Large-Scale Multi-Object Rearrangement

Eric Huang, Zhenzhong Jia, Matthew T. Mason
  • 2019 International Conference on Robotics and Automation (ICRA)
  • 2019
VIEW 2 EXCERPTS
CITES METHODS & BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-10 OF 33 REFERENCES

Similar Papers