Reinforcement learning for autonomous robot navigation.
William Ward ArmstrongBrant CoghlanDmitry O. GorodnichyPublished in: IJCNN (1999)
Keyphrases
- autonomous robot navigation
- reinforcement learning
- function approximation
- state space
- optimal policy
- learning process
- markov decision processes
- real robot
- control problems
- reinforcement learning algorithms
- machine learning
- action space
- real time
- policy search
- temporal difference learning
- learning problems
- learning algorithm
- learning capabilities
- multiscale
- perceptual aliasing
- partially observable
- dynamic programming
- temporal difference
- model free
- objective function
- database
- transfer learning
- multi agent
- dynamic environments