Visual Navigation in Real-World Indoor Environments Using End-to-End Deep Reinforcement Learning.
Jonás KulhánekErik DernerRobert BabuskaPublished in: IEEE Robotics Autom. Lett. (2021)
Keyphrases
- end to end
- indoor environments
- visual navigation
- mobile robot
- collision avoidance
- real world
- reinforcement learning
- obstacle avoidance
- path planning
- real robot
- autonomous vehicles
- ad hoc networks
- congestion control
- text localization and recognition
- visual odometry
- multipath
- wifi
- simultaneous localization and mapping
- autonomous navigation
- robotic systems
- state space
- robot control
- motion planning
- image data