A reinforcement learning with adaptive state space recruitment strategy for real autonomous mobile robots.
Toshiyuki KondoKoji ItoPublished in: IROS (2002)
Keyphrases
- state space
- reinforcement learning
- autonomous mobile robots
- markov decision processes
- reinforcement learning algorithms
- adaptive strategies
- optimal policy
- mobile robot
- markov decision process
- dynamical systems
- heuristic search
- action space
- partially observable
- dynamic programming
- machine learning
- continuous state spaces
- state abstraction
- belief state
- state variables
- particle filter
- information technology
- temporal difference
- planning problems
- object recognition
- multi agent
- model free
- reward function
- indoor environments
- learning capabilities
- dynamic environments
- real time
- reward shaping