Actor-Critic Reinforcement Learning for Control With Stability Guarantee.
Minghao HanLixian ZhangJun WangWei PanPublished in: IEEE Robotics Autom. Lett. (2020)
Keyphrases
- actor critic
- reinforcement learning
- optimal control
- lyapunov stability
- control problems
- temporal difference
- policy gradient
- function approximation
- lyapunov function
- reinforcement learning algorithms
- approximate dynamic programming
- neuro fuzzy
- gradient method
- control strategy
- adaptive control
- control policy
- action selection
- control strategies
- infinite horizon
- state space
- nonlinear systems
- natural actor critic
- dynamic programming
- learning algorithm
- model free
- average reward
- temporal difference learning
- control method
- markov decision processes
- control system
- policy gradient methods
- policy iteration
- reinforcement learning methods