Visual Navigation in Real-World Indoor Environments Using End-to-End Deep Reinforcement Learning

Jonas Kulhanek, Erik Derner, Robert Babuska

Research output: Contribution to journalArticleScientificpeer-review

8 Downloads (Pure)

Abstract

Visual navigation is essential for many applications in robotics, from manipulation, through mobile robotics to automated driving. Deep reinforcement learning (DRL) provides an elegant map-free approach integrating image processing, localization, and planning in one module, which can be trained and therefore optimized for a given environment. However, to date, DRL-based visual navigation was validated exclusively in simulation, where the simulator provides information that is not available in the real world, e.g., the robot's position or segmentation masks. This precludes the use of the learned policy on a real robot. Therefore, we present a novel approach that enables a direct deployment of the trained policy on real robots. We have designed a new powerful simulator capable of domain randomization. To facilitate the training, we propose visual auxiliary tasks and a tailored reward scheme. The policy is fine-tuned on images collected from real-world environments. We have evaluated the method on a mobile robot in a real office environment. The training took approximately 30 hours on a single GPU. In 30 navigation experiments, the robot reached a 0.3-meter neighbourhood of the goal in more than 86.7% of cases. This result makes the proposed method directly applicable to tasks like mobile manipulation.

Original languageEnglish
Pages (from-to)4345-4352
JournalIEEE Robotics and Automation Letters
Volume6
Issue number3
DOIs
Publication statusPublished - 2021

Keywords

  • Cameras
  • deep learning methods
  • Navigation
  • reinforcement learning
  • Reinforcement learning
  • Robots
  • Task analysis
  • Training
  • Vision-based navigation
  • Visualization

Fingerprint

Dive into the research topics of 'Visual Navigation in Real-World Indoor Environments Using End-to-End Deep Reinforcement Learning'. Together they form a unique fingerprint.

Cite this