Using a reinforcement learning controller to overcome simulator/environment discrepancies.
Nancy OwensTodd PetersonPublished in: SMC (2001)
Keyphrases
- reinforcement learning
- real robot
- real time
- simulation environment
- control architecture
- mobile robot
- control system
- neural network
- learning process
- reinforcement learning algorithms
- markov decision processes
- simulation tool
- test bed
- dynamic environments
- multi agent systems
- interaction model
- multi agent
- exploration strategy
- multi agent environments