DCOB: Action space for reinforcement learning of high DoF robots.
Akihiko YamaguchiJun TakamatsuTsukasa OgasawaraPublished in: Auton. Robots (2013)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- continuous state
- real valued
- mobile robot
- stochastic processes
- multi robot
- robot control
- state action
- control policies
- function approximators
- action selection
- function approximation
- reinforcement learning methods
- single agent
- markov decision problems
- continuous state spaces
- multi agent
- robotic systems
- heuristic search
- markov decision process
- learning algorithm
- real robot
- reinforcement learning algorithms
- optimal policy
- cooperative
- finite state
- path planning
- dynamic programming
- machine learning