Using Reinforcement Learning to Attenuate for Stochasticity in Robot Navigation Controllers.
James GillespieIñaki RañóNazmul H. SiddiqueJosé SantosMehdi KhamassiPublished in: SSCI (2019)
Keyphrases
- robot navigation
- reinforcement learning
- continuous state
- autonomous mobile robot
- autonomous robots
- function approximation
- scene understanding
- state space
- real time stereo
- landmark recognition
- learning algorithm
- markov decision processes
- map building
- machine learning
- optimal policy
- model free
- optimal control
- vision system
- reward function
- mobile robot
- image sequences
- computer vision