Implementation of Reinforcement Learning by transfering sub-goal policies in robot navigation.
Baris GökçeH. Levent AkinPublished in: SIU (2013)
Keyphrases
- dynamic programming
- robot navigation
- optimal policy
- reinforcement learning
- policy search
- markov decision processes
- state space
- autonomous mobile robot
- continuous state
- autonomous robots
- initially unknown
- control policies
- map building
- machine learning
- scene understanding
- topological map
- long run
- reinforcement learning algorithms
- reward function
- real time stereo
- model free
- fitted q iteration
- function approximation
- high quality
- computer vision