Reinforcement learning for landmark-based robot navigation.
Dídac BusquetsRamón López de MántarasCarles SierraPublished in: AAMAS (2002)
Keyphrases
- robot navigation
- reinforcement learning
- continuous state
- autonomous mobile robot
- autonomous robots
- scene understanding
- real time stereo
- function approximation
- landmark recognition
- state space
- learning algorithm
- robot control
- transfer learning
- markov decision processes
- model free
- initially unknown
- reward function
- action selection
- feature selection