Local Update Dynamic Policy Programming in reinforcement learning of pneumatic artificial muscle-driven humanoid hand control.
Yunduan CuiTakamitsu MatsubaraKenji SugimotoPublished in: Humanoids (2015)
Keyphrases
- learning algorithm
- reinforcement learning
- control policy
- action selection
- control policies
- control system
- control problems
- optimal policy
- degrees of freedom
- dynamically updated
- function approximation
- adaptive control
- optimal control
- reinforcement learning algorithms
- force control
- model free
- state space
- programming language
- dynamic environments
- partially observable markov decision processes
- markov decision process
- mathematical model
- partially observable environments
- long run
- reinforcement learning problems
- continuous state
- function approximators
- robot control
- humanoid robot
- partially observable
- control method
- motion planning
- control law