FollowNet: Robot Navigation by Following Natural Language Directions with Deep Reinforcement Learning.
Pararth ShahMarek FiserAleksandra FaustJ. Chase KewDilek Hakkani-TürPublished in: CoRR (2018)
Keyphrases
- robot navigation
- reinforcement learning
- natural language
- continuous state
- autonomous robots
- autonomous mobile robot
- function approximation
- scene understanding
- machine learning
- real time stereo
- state space
- markov decision processes
- initially unknown
- map building
- landmark recognition
- robot localization
- dynamic programming
- model free
- topological map
- learning algorithm
- function approximators
- action selection
- optimal policy
- high quality