Stable and Safe Reinforcement Learning via a Barrier-Lyapunov Actor-Critic Approach.
Liqun ZhaoKonstantinos GatsisAntonis PapachristodoulouPublished in: CDC (2023)
Keyphrases
- actor critic
- reinforcement learning
- optimal control
- temporal difference
- policy gradient
- reinforcement learning algorithms
- approximate dynamic programming
- function approximation
- gradient method
- adaptive control
- neuro fuzzy
- dynamical systems
- control law
- lyapunov function
- policy gradient methods
- state space
- policy iteration
- model free
- stability analysis
- control scheme
- closed loop
- nonlinear systems
- dynamic programming
- natural actor critic
- machine learning
- markov decision processes
- reinforcement learning methods
- action selection
- linear program
- control problems
- step size
- sufficient conditions
- mobile robot