Target-driven visual navigation in indoor scenes using deep reinforcement learning

@article{Zhu2017TargetdrivenVN,
  title={Target-driven visual navigation in indoor scenes using deep reinforcement learning},
  author={Yuke Zhu and R. Mottaghi and Eric Kolve and Joseph J. Lim and A. Gupta and Li Fei-Fei and Ali Farhadi},
  journal={2017 IEEE International Conference on Robotics and Automation (ICRA)},
  year={2017},
  pages={3357-3364}
}
  • Yuke Zhu, R. Mottaghi, +4 authors Ali Farhadi
  • Published 2017
  • Computer Science, Engineering
  • 2017 IEEE International Conference on Robotics and Automation (ICRA)
  • Two less addressed issues of deep reinforcement learning are (1) lack of generalization capability to new goals, and (2) data inefficiency, i.e., the model requires several (and often costly) episodes of trial and error to converge, which makes it impractical to be applied to real-world scenarios. In this paper, we address these two issues and apply our model to target-driven visual navigation. To address the first issue, we propose an actor-critic model whose policy is a function of the goal… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 606 CITATIONS, ESTIMATED 97% COVERAGE

    Vision-based Navigation Using Deep Reinforcement Learning

    VIEW 10 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    Visual Navigation using Deep Reinforcement Learning

    VIEW 1 EXCERPT
    CITES METHODS

    A deep Q network for robotic planning from image

    VIEW 1 EXCERPT
    CITES METHODS

    Reinforcement Learning based Visual Navigation with Information-Theoretic Regularization

    VIEW 10 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Cross-View Policy Learning for Street Navigation

    VIEW 3 EXCERPTS
    CITES BACKGROUND

    Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation

    VIEW 2 EXCERPTS
    CITES BACKGROUND

    Vision-Language Navigation Policy Learning and Adaptation.

    VIEW 1 EXCERPT
    CITES BACKGROUND

    FILTER CITATIONS BY YEAR

    2016
    2020

    CITATION STATISTICS

    • 61 Highly Influenced Citations

    • Averaged 180 Citations per year from 2018 through 2020

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 55 REFERENCES

    End-to-End Training of Deep Visuomotor Policies

    VIEW 1 EXCERPT

    ViZDoom: A Doom-based AI research platform for visual reinforcement learning

    VIEW 2 EXCERPTS

    Policy Distillation

    VIEW 1 EXCERPT

    View-based Maps

    VirtualWorlds as Proxy for Multi-object Tracking Analysis

    VIEW 1 EXCERPT