Vision-based Navigation Using Deep Reinforcement Learning

  title={Vision-based Navigation Using Deep Reinforcement Learning},
  author={Jon{\'a}s Kulh{\'a}nek and Erik Derner and T. D. Bruin and R. Babu{\vs}ka},
  journal={2019 European Conference on Mobile Robots (ECMR)},
  • Jonás Kulhánek, Erik Derner, +1 author R. Babuška
  • Published 2019
  • Computer Science, Engineering, Mathematics
  • 2019 European Conference on Mobile Robots (ECMR)
  • Deep reinforcement learning (RL) has been successfully applied to a variety of game-like environments. However, the application of deep RL to visual navigation with realistic environments is a challenging task. We propose a novel learning architecture capable of navigating an agent, e.g. a mobile robot, to a target given by an image. To achieve this, we have extended the batched A2C algorithm with auxiliary tasks designed to improve visual navigation performance. We propose three additional… CONTINUE READING
    10 Citations

    Figures, Tables, and Topics from this paper

    Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation
    • Juncheng Li, X. Wang, +4 authors William Yang Wang
    • Computer Science
    • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
    • 2020
    • 4
    • PDF
    Visual Navigation in Real-World Indoor Environments Using End-to-End Deep Reinforcement Learning
    • PDF
    Object recognition for UAV navigation in complex environment


    Target-driven visual navigation in indoor scenes using deep reinforcement learning
    • 745
    • Highly Influential
    • PDF
    One-Shot Reinforcement Learning for Robot Navigation with Interactive Replay
    • 37
    • PDF
    Learning to Navigate in Complex Environments
    • 516
    • PDF
    Reinforcement Learning with Unsupervised Auxiliary Tasks
    • 688
    • Highly Influential
    • PDF
    Deep auto-encoder neural networks in reinforcement learning
    • 226
    • PDF
    Asynchronous Methods for Deep Reinforcement Learning
    • 3,783
    • Highly Influential
    • PDF
    Playing Atari with Deep Reinforcement Learning
    • 4,734
    • Highly Influential
    • PDF
    Learning state representation for deep actor-critic control
    • 38
    • PDF
    Dueling Network Architectures for Deep Reinforcement Learning
    • 1,380
    • PDF
    AI2-THOR: An Interactive 3D Environment for Visual AI
    • 233
    • Highly Influential
    • PDF