NavRep: Unsupervised Representations for Reinforcement Learning of Robot Navigation in Dynamic Human Environments.
Daniel DugasJuan I. NietoRoland SiegwartJen Jen ChungPublished in: CoRR (2020)
Keyphrases
- robot navigation
- autonomous robots
- reinforcement learning
- continuous state
- dynamic environments
- autonomous mobile robot
- real time stereo
- mobile robot
- supervised learning
- scene understanding
- unsupervised learning
- landmark recognition
- robot control
- semi supervised
- machine learning
- real time
- topological map
- multi class
- initially unknown